[House Hearing, 111 Congress]
[From the U.S. Government Publishing Office]


 
       THE STATE OF RESEARCH INFRASTRUCTURE AT U.S. UNIVERSITIES

=======================================================================

                                HEARING

                               BEFORE THE

             SUBCOMMITTEE ON RESEARCH AND SCIENCE EDUCATION

                  COMMITTEE ON SCIENCE AND TECHNOLOGY
                        HOUSE OF REPRESENTATIVES

                     ONE HUNDRED ELEVENTH CONGRESS

                             SECOND SESSION

                               __________

                           FEBRUARY 23, 2010

                               __________

                           Serial No. 111-77

                               __________

     Printed for the use of the Committee on Science and Technology


     Available via the World Wide Web: http://www.science.house.gov




                  U.S. GOVERNMENT PRINTING OFFICE
55-835                    WASHINGTON : 2010
-----------------------------------------------------------------------
For Sale by the Superintendent of Documents, U.S. Government Printing Office
Internet: bookstore.gpo.gov  Phone: toll free (866) 512-1800; (202) 512ï¿½091800  
Fax: (202) 512ï¿½092104 Mail: Stop IDCC, Washington, DC 20402ï¿½090001
                                 ______

                  COMMITTEE ON SCIENCE AND TECHNOLOGY

                   HON. BART GORDON, Tennessee, Chair
JERRY F. COSTELLO, Illinois          RALPH M. HALL, Texas
EDDIE BERNICE JOHNSON, Texas         F. JAMES SENSENBRENNER JR., 
LYNN C. WOOLSEY, California              Wisconsin
DAVID WU, Oregon                     LAMAR S. SMITH, Texas
BRIAN BAIRD, Washington              DANA ROHRABACHER, California
BRAD MILLER, North Carolina          ROSCOE G. BARTLETT, Maryland
DANIEL LIPINSKI, Illinois            VERNON J. EHLERS, Michigan
GABRIELLE GIFFORDS, Arizona          FRANK D. LUCAS, Oklahoma
DONNA F. EDWARDS, Maryland           JUDY BIGGERT, Illinois
MARCIA L. FUDGE, Ohio                W. TODD AKIN, Missouri
BEN R. LUJAN, New Mexico             RANDY NEUGEBAUER, Texas
PAUL D. TONKO, New York              BOB INGLIS, South Carolina
JOHN GARAMENDI, California           MICHAEL T. McCAUL, Texas
STEVEN R. ROTHMAN, New Jersey        MARIO DIAZ-BALART, Florida
JIM MATHESON, Utah                   BRIAN P. BILBRAY, California
LINCOLN DAVIS, Tennessee             ADRIAN SMITH, Nebraska
BEN CHANDLER, Kentucky               PAUL C. BROUN, Georgia
RUSS CARNAHAN, Missouri              PETE OLSON, Texas
BARON P. HILL, Indiana
HARRY E. MITCHELL, Arizona
CHARLES A. WILSON, Ohio
KATHLEEN DAHLKEMPER, Pennsylvania
ALAN GRAYSON, Florida
SUZANNE M. KOSMAS, Florida
GARY C. PETERS, Michigan
VACANCY
                                 ------                                

             Subcommittee on Research and Science Education

                 HON. DANIEL LIPINSKI, Illinois, Chair
EDDIE BERNICE JOHNSON, Texas         VERNON J. EHLERS, Michigan
BRIAN BAIRD, Washington              RANDY NEUGEBAUER, Texas
MARCIA L. FUDGE, Ohio                BOB INGLIS, South Carolina
PAUL D. TONKO, New York              BRIAN P. BILBRAY, California
RUSS CARNAHAN, Missouri                  
VACANCY                                  
BART GORDON, Tennessee               RALPH M. HALL, Texas
               DAHLIA SOKOLOV Subcommittee Staff Director
            MARCY GALLO Democratic Professional Staff Member
           MELE WILLIAMS Republican Professional Staff Member
                    BESS CAUGHRAN Research Assistant


                            C O N T E N T S

                           February 23, 2010

                                                                   Page
Witness List.....................................................     2

Hearing Charter..................................................     3

                           Opening Statements

Statement by Representative Daniel Lipinski, Chairman, 
  Subcommittee on Research and Science Education, Committee on 
  Science and Technology, U.S. House of Representatives..........     7
    Written Statement............................................     8

Statement by Representative Vernon J. Ehlers, Minority Ranking 
  Member, Subcommittee on Research and Science Education, 
  Committee on Science and Technology, U.S. House of 
  Representatives................................................     9
    Written Statement............................................    13

Prepared Statement by Representative Eddie Bernice Johnson, 
  Member, Subcommittee on Research and Science Education, 
  Committee on Science and Technology, U.S. House of 
  Representatives................................................    13

                               Witnesses:

Dr. Leslie P. Tolbert, Vice President for Research, Graduate 
  Studies and Economic Development, University of Arizona
    Oral Statement...............................................    15
    Written Statement............................................    17
    Biography....................................................    26

Mr. Albert G. Horvath, Senior Vice President for Finance and 
  Business, Pennsylvania State University, and Chair, Board of 
  Directors, Counsel on Government Relations
    Oral Statement...............................................    27
    Written Statement............................................    28
    Biography....................................................    47

Dr. John R. Raymond, Vice President for Academic Affairs and 
  Provost, Medical University of South Carolina, and Chair, State 
  of South Carolina EPSCoR Committee
    Oral Statement...............................................    48
    Written Statement............................................    49
    Biography....................................................    55

Dr. Thom H. Dunning, Jr., Director of the National Center for 
  Supercomputing Applications, University of Illinois at Urbana-
  Champaign
    Oral Statement...............................................    56
    Written Statement............................................    58
    Biography....................................................    69


       THE STATE OF RESEARCH INFRASTRUCTURE AT U.S. UNIVERSITIES

                              ----------                              


                       TUESDAY, FEBRUARY 23, 2010

                  House of Representatives,
     Subcommittee on Research and Science Education
                        Committee on Science and Technology
                                                    Washington, DC.
    The Subcommittee met, pursuant to call, at 2:03 p.m., in 
Room 2318 of the Rayburn House Office Building, Hon. Daniel 
Lipinski [Chairman of the Subcommittee] presiding.


                            hearing charter

                     U.S. HOUSE OF REPRESENTATIVES

                  COMMITTEE ON SCIENCE AND TECHNOLOGY

                  SUBCOMMITTEE ON RESEARCH AND SCIENCE

                               EDUCATION

       The State of Research Infrastructure at U.S. Universities

                       tuesday, february 23, 2010
                          2:00 p.m.-4:00 p.m.
                   2318 rayburn house office building

1. Purpose

    The purpose of this hearing is to examine the research and research 
training infrastructure of our universities and colleges, including 
research facilities, and cyberinfrastructure capabilities, the capacity 
of the research infrastructure to meet the needs of U.S. scientists and 
engineers now and in the future, and the appropriate role of the 
Federal government in sustaining such infrastructure.

2. Witnesses:

          Dr. Leslie Tolbert, Vice President for Research, 
        Graduate Studies and Economic Development, University of 
        Arizona

          Mr. Albert Horvath, Senior Vice President for Finance 
        and Business, Pennsylvania State University

          Dr. John R. Raymond, Vice President for Academic 
        Affairs and Provost, Medical University of South Carolina, and 
        Chair, State of South Carolina EPSCoR Committee

          Dr. Thom Dunning, Director of the National Center for 
        Supercomputing Applications, University of Illinois at Urbana-
        Champaign

3. Overarching Questions:

  What is the state of the nation's academic research 
facilities? Are current academic research facilities keeping U.S. 
scientists and engineers competitive with their international 
counterparts and are they allowing for cutting edge science? How are 
universities and colleges maintaining and improving their research 
facilities? How has the economic climate affected short-term and long-
term planning and investments in academic research facilities?

  What is the status of the nation's cyberinfrastructure? Do 
our research and education networks have the capacity to support 
computational, storage, data transfer and scientific exchange needs 
that have become critical to performing innovative research? How are 
universities and colleges investing in their own cyberinfrastructure? 
How are universities partnering with state and local governments as 
well as the private sector to build regional cyberinfrastructure 
capabilities?

  What is the appropriate role of the Federal government in 
supporting the research infrastructure of our universities and 
colleges? How do Federal agencies such as the National Science 
Foundation support research infrastructure that benefits the science 
and engineering enterprise? Given the trade-off between support for 
research and the support of research facilities should NSF revive their 
Academic Research Infrastructure Program? What other options, beyond 
targeted programs, are there for Federal science agencies to support 
academic research infrastructure?

4. Background

University Research Infrastructure

    Since 1988, NSF has conducted a biennial survey on the status of 
research facilities at academic institutions, nonprofit biomedical 
research organizations and university hospitals. The survey currently 
includes data on: the amount of research space, the condition of 
research facilities, current expenditures and plans for new 
construction as well as the renovation of research facilities, sources 
of funds for construction and renovation, and information technology 
capabilities.
    According to the latest NSF survey,\1\ 77 percent of the 
respondents rated the condition of their research space as satisfactory 
or superior with the remainder indicating that their research space 
needed to be renovated or replaced. The survey also showed that 
academic institutions spent $6.1 billion on new construction and $2.4 
billion on the repair and renovation of research facilities, but 
deferred $10.2 billion in new construction projects and $3.5 billion in 
renovation projects. Despite deferred investments, the amount of 
research space at academic institutions has steadily increased to 192 
million square feet in 2007, although the rate of increase has slowed 
to 3.7 percent, down from its peak of 11 percent between 2001 and 2003.
---------------------------------------------------------------------------
    \1\ http://www.nsf.gov/statistics/nsf07325/
---------------------------------------------------------------------------
    Academic institutions fund their capital investments through a 
combination of sources: the Federal government, state and local 
governments, and institutional funds, which include endowments, private 
donations, and facilities and administration (F&A) costs recovered from 
the Federal government. The Federal share of these capital investments 
is generally about five percent, with the state/local governments 
accounting for 22 percent, and the institutions themselves contributing 
72 percent. As just noted, the institutional share does include F&A 
costs reimbursed by the Federal government as part of Federal contracts 
and grants, primarily research grants. The reimbursed funds are used 
for such activities as operation and maintenance of research 
facilities, library expenses, department administration, including 
secretaries, academic deans, and grant compliance officers. However, 
according to a 2000 RAND study,\2\ the true F&A costs incurred by an 
institution are higher than the rate for which they are reimbursed and 
analyses indicate that universities are recouping between 70 to 90 
percent of the amount they are actually spending on facilities and 
administration.
---------------------------------------------------------------------------
    \2\ http://www.rand.org/pubs/monograph-reports/MIR1135-
1/

---------------------------------------------------------------------------
Cyberinfrastructure

    Advances in information technology have changed the way research is 
conducted. In 2005, NSF created the Office of Cyberinfrastructure (OCT) 
to ensure a comprehensive vision and set of investments in the 
research, development, acquisition, and operation of 
cyberinfrastructure across NSF's research directorates. 
Cyberinfrastructure, which consists of computing systems, data storage 
systems, data repositories, advanced instruments, and the networks and 
software that link these systems, has become increasingly important to 
all science and engineering disciplines. OCI requested a budget of $228 
million in FY 2011, a 6.4 percent increase from FY 2010, with the 
largest investment proposed for the development of petascale computing 
capabilities.
    NSF's recent Science and Engineering Indicators report\3\ shows 
that all institutions of higher education have access to the internet, 
which was not the case earlier in the decade, but the bandwidth 
capability or speed of internet connection varied across institution 
type. The overwhelming majority (83 percent) of institutions with a 
bandwidth of at least 1 gigabit per second were doctoral degree 
granting institutions, and all but one institution with a bandwidth 
greater than 2.4 gigabits per second granted doctoral degrees. Despite 
the current differences in capabilities, data from NSF indicates that 
all colleges and universities are investing heavily in the expansion of 
their networks and are improving wireless campus coverage as well as 
their external and internal network speeds.
---------------------------------------------------------------------------
    \3\ http://www.nsf.gov/statistics/seind10/

---------------------------------------------------------------------------
NSF's Academic Research Infrastructure Program

    The Academic Research Infrastructure (ARI) program was originally 
authorized by the Science and Technology Committee in 1988, with 
funding authorized through 1993. The authorization level grew from $80 
million in 1989 to $250 million in 1993. The original ARI program 
consisted of two components: support for the acquisition or development 
of major research instrumentation and support for the improvement of 
research and research training facilities.
    ARI was included in appropriations bills from 1990 until 1996. It 
was initially funded at $20 million, and rose steadily to $100 million 
with an anomalous peak of $250 million in 1995. Beginning in 1997, NSF 
continued the instrumentation part of ARI only, and renamed it the 
Major Research Instrumentation (MRI) Program. The funding level for MRI 
in 1997 was $50 million, half the level the full ART program received 
the year before. Today, it receives approximately $100 million annually 
with a FY 2011 budget request of $90 million. MRI also received $300 
million in the Recovery Act, which helped NSF fill in much of the 
backlog in demand from universities.
    The long defunct facilities portion of the old ART program received 
$200 million in the Recovery Act. NSF stood up a revised version of the 
program, the Academic Research Infrastructure Program: Recovery and 
Reinvestment (ARI-R2), that does not require cost sharing 
and goes beyond physical research facilities, allowing for the 
modernization of virtual research space. Last August, NSF received 495 
applications for funding under the ARI-R2 program, proposing 
a total of $1.2 billion in renovations. NSF plans to award 125 grants 
between February and September in three size categories: $250,000-$2 
million, $2 million-$5 million, and $5 million-$10 million. According 
to NSF, the vast majority. of awards will fall into the $250,000 to $2 
million range. Additionally, nearly half of the awards (46 percent) 
will go to doctoral degree granting institutions, with the remaining 
going to a variety of master's degree granting institutions, 
undergraduate institutions, minority serving institutions and non-
profit research organizations. The overall success rate of 25 percent 
is similar to the Foundation-wide success rate for its competitive 
awards.

NSF Support for Research Infrastructure Broadly

    In addition to supporting cutting edge science through research 
grants, NSF invests in the infrastructure that enables such research. 
Approximately 24 percent ($1.8 billion) of NSF's FY 2011 budget is 
devoted to research infrastructure. These infrastructure investments 
are generally large, multi-user facilities, distributed instrumentation 
networks, or large pieces of equipment such as telescopes, research 
vessels, or accelerators that benefit an entire scientific discipline 
and could not be achieved without significant Federal support. For 
example, the Ocean Observatories Initiative, currently under 
construction with funding from the Major Research Equipment and 
Facilities (MREFC) account, will create a network of sensors for the 
continuous and real-time measurement of the physical, chemical, 
geological and biological variables of the ocean and seafloor.
    In addition to these targeted large-scale investments, NSF also 
supports the development of university research infrastructure through 
the Experimental Program to Stimulate Competitive Research (EPSCoR) 
program. EPSCoR was created in 1978 to build research capacity in 
States with few research intensive universities; in order to be 
eligible a state must receive less than 0.75 percent of the total NSF 
funding awarded in the previous three-year period. The intent of the 
program is to improve a state's competitiveness for R&D funding 
primarily by supporting sustainable research infrastructure 
improvements across the states' academic institutions. NSF has 
requested $154 million for the program in FY 2011, a five percent 
increase from FY 2010. The success of NSF's EPSCoR program in the 1980s 
resulted in the creation of six other EPSCoR-like programs within DOE, 
DOD, NIH, NASA, EPA, and USDA. In FY 2008, these programs invested a 
total of $419 million across the approximately 25 EPSCoR-eligible 
states.

5. Questions for Witnesses

Dr. Leslie Tolbert

    1. How does the University of Arizona plan for its research 
infrastructure needs, including its research facilities? What is the 
current state the University of Arizona's research infrastructure and 
its plans for the next 5-10 years? How is the University of Arizona 
partnering with state and local governments as well as the private 
sector to improve the region's research infrastructure and 
capabilities?

    2. What federal funds currently support the University of Arizona's 
research infrastructure, including research facilities? Please include 
a description of all sources of funding for your research facilities, 
including indirect costs reimbursed from federal research grants. What 
are your unmet research infrastructure needs? Would you support funding 
for the Academic Research Infrastructure Program if it meant decreasing 
NSF's research budget by an equivalent amount? Are there other options 
beyond targeted programs for Federal science agencies to support the 
research infrastructure of our universities?

Mr. Albert Horvath

    1. Please describe how research infrastructure is financed at 
Pennsylvania State University, including the financing of research 
facilities, cyberinfrastructure and other investments in the 
university's research capabilities? What federal funds currently 
support Penn State's research infrastructure, including research 
facilities? Please include a description of all sources of funding for 
your research facilities, including indirect costs reimbursed from 
federal research grants. What are your unmet research infrastructure 
needs?

    2. How is Penn State partnering with state and local governments as 
well as the private sector to improve the region's research 
infrastructure and capabilities?

    3. Would you support funding for the Academic Research 
Infrastructure Program if it meant decreasing NSF's research budget by 
an equivalent amount? Are there other options beyond targeted programs 
for Federal science agencies to support the research infrastructure of 
our universities?

Dr. John R. Raymond

    1. Please describe the current National Science Foundation EPSCoR 
grant awarded to South Carolina. What role have EPSCoR funds played in 
facilitating partnerships with state and local governments as well as 
the private sector to improve the region's research infrastructure and 
capabilities? How have EPSCoR funds been leveraged across institutions 
to improve the region's cyberinfrastructure capabilities? Specifically, 
how have EPSCoR funds been used by the Medical University of South 
Carolina?

    2. Please describe the state of Medical University of South 
Carolina's research infrastructure, including its research facilities. 
What are your unmet research infrastructure needs?

    3. Do you have any specific recommendations on how to improve the 
EPSCoR program? Are there other options beyond targeted programs for 
Federal science agencies to support the research infrastructure of our 
universities?

Dr. Thom Dunning

    1. Please describe the state of the University of Illinois's 
cyberinfrastructure, including the Blue Waters project. How is the 
University of Illinois partnering with state and local governments as 
well as the private sector to build regional cyberinfrastructure 
capabilities?

    2. In your opinion, as the lead of one of six task forces 
established by NSF's Advisory Committee for Cyberinfrastructure to 
address long-term cyberinfrastructure issues, what is the state of the 
Nation's cyberinfrastructure? Do our research and education networks 
have the capacity to support computational, storage, data transfer and 
other scientific exchange needed to perform innovative research? Are we 
appropriately prioritizing our investments in cyberinfrastructure? 
What, if any, critical investments or opportunities are we missing?
    Chairman Lipinski. This hearing will now come to order.
    Good afternoon. I want to welcome you to the Research and 
Science Education Subcommittee hearing on the state of our 
universities' infrastructure for research and research 
training. This is one in a series of hearings and roundtables 
that we are holding as this Subcommittee works on the bill to 
reauthorize the National Science Foundation and the Committee 
works on the reauthorization of the America COMPETES Act. Our 
focus on this legislation is a direct acknowledgement of the 
fact that America's science and technology enterprise underpins 
the long-term economic competitiveness of our country, and we 
need to do whatever we can to make sure we maintain that.
    Over the past 60 years, a great number of our societal and 
economic benefits have come out of the highly successful 
partnership between the Federal Government and our Nation's 
colleges and universities. Not only do these institutions train 
the workforce needed in a modern economy, but they also conduct 
the research that generates new knowledge and technologies. It 
is a testament to the productivity of this arrangement that 80 
percent of the National Science Foundation research dollars go 
to academic institutions.
    But successful R&D takes more than intellectual freedom and 
grant funding. You also need state-of-the-art lab space, 
networks, instruments and computing facilities, and I have 
heard some concerns from the academic community that this 
infrastructure is being neglected. Public institutions 
especially are suffering as the recession has eroded state 
support. I am worried that unless we actively modernize our R&D 
facilities that we could not only be spending federal research 
dollars inefficiently, but that we could lose our position as 
scientific leaders, finding it harder to attract top scientists 
and engineers.
    Our competitors are investing in all aspects of their R&D 
ecosystems. Only a decade ago, if you asked an exceptional 
Chinese graduate student in science or engineering whether they 
would rather return home or stay and become an American 
citizen, nearly all of them would have chosen the latter. But 
that is no longer the case, with the best students increasingly 
being lured back home. At least part of the reason for this is 
the new availability of cutting-edge facilities and support 
they need to succeed as researchers.
    Today we want to examine the state of our universities' 
research infrastructure and to consider the federal role in 
supporting this infrastructure, in particular the appropriate 
balance between investing in the research itself and investing 
in the infrastructure that underlies and supports both research 
activities and workforce training.
    Currently, universities maintain and upgrade their own 
campus-based facilities with funding from a variety of sources. 
Federal agencies such as the NSF directly or indirectly support 
some of this infrastructure, but their primary mission is to 
support research and multi-user facilities that benefit the 
scientific enterprise and society broadly. Historically, 
however, the NSF has at times funded merit-based academic 
research infrastructure. For example, in the 1960s and 1970s 
the NSF ran a laboratory development program, an institutional 
science grant program, and a development program for University 
Centers of Excellence. In the mid-1980s, this Committee 
systematically examined the issue, beginning by requiring the 
NSF to prepare biannual reports on the research facilities 
needs of universities, and ultimately passing the Academic 
Research Facilities Modernization Act. This Act led to both the 
NSF's Major Research Instrumentation Program and the Academic 
Research Infrastructure Program.
    But apart from one-time funding in the stimulus bill last 
year, federal programs to modernize scientific infrastructure 
have languished in recent years. Perhaps as a result, the 2005 
Survey of Science and Engineering Research Facilities found 
that academic institutions were deferring $3.5 billion in 
needed renovation projects.
    During today's hearing, I want to hear our witnesses' 
thoughts on whether they think the NSF should once again 
directly invest in research infrastructure for universities. 
Obviously, even with increases in NSF funding, tradeoffs would 
have to be made. I also hope to learn about how academic 
institutions are currently leveraging federal investments to 
improve the research capacity of their institutions. I would 
also like to hear our witnesses' ideas on how best to ensure 
that our research infrastructure keeps up with both the 
frontiers of the science and our international competitors. 
Finally, I am interested in learning about the opportunities to 
expand our research and educational capabilities through 
growing our cyberinfrastructure.
    I want to thank the witnesses for their flexibility in the 
rescheduling of today's hearing. The Committee already had a 
very full calendar over the next few weeks, and with hearings 
postponed by this month's record snowfall, some changes had to 
be made, and I want to thank the witnesses for their 
flexibility. Thank you for being here this afternoon and I look 
forward to your testimony.
    Before I move on and recognize Dr. Ehlers, I just want to 
say that I am going to--with the announcement that Dr. Ehlers 
made that he will not be running again for reelection, I want 
to say how much I will miss having him here. We only had about 
a year here on this Subcommittee together but he has always 
provided not only his knowledge but his passion for what we are 
working on, and we have worked very well together. I know that 
things don't always run very well up here on Capitol Hill but 
Vern has made things run very well, and I think for the 
betterment of this Committee and the betterment of our country, 
and I look forward to the next ten more months that we have to 
work together on this Committee.
    [The prepared statement of Chairman Lipinski follows:]

             Prepared Statement of Chairman Daniel Lipinski

    Good afternoon and welcome to this Research and Science Education 
Subcommittee hearing on the state of our universities' infrastructure 
for research AND research training. This is one in a series of hearings 
and roundtables that we are holding as this subcommittee works on the 
bill reauthorizing the National Science Foundation and the committee 
works on the reauthorization of the America COMPETES Act. Our focus on 
this legislation is a direct acknowledgement of the fact that America's 
science and technology enterprise underpins the long-term economic 
competitiveness of our country.
    Over the past 60 years, a great number of societal and economic 
benefits have come out of the highly successful partnership between the 
Federal government and our Nation's colleges and universities. Not only 
do these institutions train the workforce needed in a modern economy, 
but they also conduct the research that generates new knowledge and 
technologies. It is a testament to the productivity of this arrangement 
that eighty percent of the National Science Foundation's research 
dollars go to academic institutions.
    But successful R&D takes more than intellectual freedom and grant 
funding. You also need state-of-the-art lab space, networks, 
instruments, and computing facilities, and I have heard some concerns 
from the academic community that this infrastructure is being 
neglected. Public institutions especially are suffering as the 
recession has eroded state support. I am worried that unless we 
actively modernize our R&D facilities that we could not only be 
spending Federal research dollars inefficiently, but that we could lose 
our position as scientific leaders, finding it harder to attract top 
scientists and engineers.
    Our competitors are investing in all aspects of their R&D 
ecosystems. Only a decade ago, if you asked an exceptional Chinese 
graduate student in science or engineering whether they would rather 
return home or stay and become an American citizen, nearly all of them 
would have chosen the latter. But that is no longer the case, with the 
best students increasingly being lured back home. At least part of the 
reason for this is the new availability of cutting-edge facilities and 
support they need to succeed as researchers.
    Today we want to examine the state of our universities' research 
infrastructure and to consider the Federal role in supporting this 
infrastructure, in particular the appropriate balance between investing 
in the research itself and investing in the infrastructure that 
underlies and supports both research activities and workforce training.
    Currently, universities maintain and upgrade their own campus-based 
facilities with funding from a variety of sources. Federal agencies 
such as the NSF directly or indirectly support some of this 
infrastructure, but their primary mission is to support research and 
multi-user facilities that benefit the scientific enterprise and 
society broadly. Historically, however, the NSF has at times funded 
merit-based academic research infrastructure. For example, in the 1960s 
and 70s the NSF ran a laboratory development program, an institutional 
science grant program, and a development program for university centers 
of excellence.
    In the mid-80s, this Committee systematically examined the issue, 
beginning by requiring the NSF to prepare biannual reports on the 
research facilities needs of universities, and ultimately passing the 
Academic Research Facilities Modernization Act. This Act led to both 
the NSF's Major Research Instrumentation Program and the Academic 
Research Infrastructure Program.
    But, apart from one-time funding in the stimulus bill last year, 
Federal programs to modernize scientific infrastructure have languished 
in recent years. Perhaps as a result, the 2005 Survey of Science and 
Engineering Research facilities found that academic institutions were 
deferring $3.5 billion in needed renovation projects.
    During today's hearing, I want to hear our witnesses thoughts on 
whether they think the NSF should once again directly invest in 
research infrastructure for universities. Obviously, even with 
increases in NSF funding, trade-offs will have to be made. I also hope 
to learn about how academic institutions are currently leveraging 
Federal investments to improve the research capacity of their 
institutions. I would also like to hear our witnesses' ideas on how 
best to ensure that our research infrastructure keeps up with both the 
frontiers of the science and our international competitors. Finally, I 
am interested in learning about the opportunities to expand our 
research and educational capabilities through growing our 
cyberinfrastructure.
    I want to thank the witnesses for their flexibility in the 
rescheduling of today's hearing. The Committee already had a very full 
calendar over the next few weeks, and with hearings postponed by this 
month's record snowfall some changes had to be made. I thank the 
witnesses for being here this afternoon and I look forward to your 
testimony.

    Chairman Lipinski. I will now recognize Dr. Ehlers for his 
opening statement.
    Mr. Ehlers. Thank you, Mr. Chairman. I appreciate those 
kind words, and I am impressed now that we have come back 
together after all the snow how many of my colleagues 
appreciate me. It is too bad they didn't say that before I 
decided not to run again. But at any rate, I appreciate your 
comments, and I just think, I have been here 16 years, it will 
be almost 17 by the time I leave, and I think that is a good 
long time and it is time for someone else to carry the banner, 
but I do hope we can get more scientists to run for the 
Congress. I have been told that I was the first research 
physicist ever to be elected to the Congress. I can't vouch for 
the accuracy of that but there is not much evidence to say that 
I wasn't. But at any rate, I hope we get a lot more scientists 
interested in the political arena because it is desperately 
needed.
    Thank you for holding this hearing too. It is a very 
important issue. And now that I am officially achieving old 
fogey status, I thought I might give a little historical 
perspective on research institutions, and I want to assure you, 
I was not born 200 years ago but that is roughly where I will 
start, because back then at the time of Newton and before there 
was no such thing as a government grant; occasionally a very 
bright individual would get support from the local king, 
prince, duke, noble, whatever, provided, of course, he also 
devoted some energy and time to developing better instruments 
of warfare. But that was a very informal arrangement for many 
years.
    That continued well into the 19th century, but that is when 
things began to change somewhat, and the first person I think 
who really has led to the development of modern science was Mr. 
Faraday, and there is a marvelous biography of his that is out 
there, so I recommend everyone to read that, but an amazing 
man. When he got out of school, he went to work in a bookstore 
selling books, and part of his pay was to have a room up in the 
attic so he literally lived in the bookstore, and proceeded to 
read virtually all the books in the bookstore. So he was self-
educated. He did not have any higher education. Then he 
discovered scientific books and read them and devoured them and 
started doing some experiments on his own in the bookstore at 
night and made some interesting discoveries. For example, he 
discovered the effect that led to the generation of electricity 
by rotating electrical coils in magnetic fields, and he 
developed very good scientific techniques. He eventually became 
a member of the Academy of Sciences equivalent in the British 
Empire, and he popularized science for the people by scheduling 
special lectures, I think it was every other week or something 
like that, and the public turned out in full force for them. It 
was as popular to go to the science lectures as it was to go to 
the symphony, and was well respected. Unfortunately, we have 
slipped from that. But he really educated a lot of laymen about 
science. He also worked on war efforts. He developed an oatmeal 
that would last and stay fresh much longer than any he ever had 
before and he was awarded medals by the British Navy because 
they were out at sea for months and so it was a problem to keep 
the food fresh. So he went into many different fields but his 
primary contributions were to electricity.
    The lectures he established were a real effort to educate 
the public and became very popular. But one time he invited Mr. 
Ohm to speak, and we are all familiar with the term ohm because 
that is the unit of resistance and electricity is named after 
him. So he was invited to give a lecture, and when they went 
down the stairs to enter the lecture room, and Faraday was 
going first because he was going to introduce Mr. Ohm, and when 
Mr. Ohm entered the room, looked around and saw 200 people, he 
panicked. He was a bit of an introvert. He panicked and ran 
back up the stairs and left. So Faraday wrote a little note. 
Being the scientist that he was, he had to record all the data, 
so there was a little note in his data book, ``When escorting 
speakers to the auditorium, be sure to follow them and not lead 
them.'' So he was a very good observer.
    Well, that is enough of history. Well, let me add just 
another bit that is more recent and I am more familiar with 
because I did some of my research using the cyclotron that 
resulted from that. E.O. Lawrence was hired by the president of 
the University of California at Berkeley to teach in the 
physics department and he was hired primarily because someone 
said that E.O. Lawrence is likely to get a Nobel Prize at some 
point. He was very bright. And the president said he wanted a 
Nobel Prize winner on the Berkeley campus so he offered him a 
job. They now have more Nobel Prize winners there than any 
other university, I believe. But in any event, E.O. Lawrence 
invented the cyclotron. It was small enough to sit on a kitchen 
chair. In fact, I have seen that original cyclotron and the 
kitchen chair. And then he realized he could scale up but where 
would he ever get a magnet that was big enough, and he 
discovered that AT&T, which had planned to send radio waves 
across the ocean, the Pacific ocean, a very long distance, and 
they were going to send those so that they could get telegrams 
across oceans, and they had the magnet and things just didn't 
work out right and then they started laying undersea cables so 
the magnet was in Palo Alto, and E.O. Lawrence heard about it, 
went and saw it, persuaded AT&T to donate it to the University 
of California, and that became the 60-inch cyclotron at 
Berkeley, which was the biggest accelerator in the world for 
quite a few years and enabled us to learn enough about nuclear 
physics that we could develop the ultimate weapon that ended 
World War II.
    Well, that is a lot of history, but the reason I am 
bringing that up, because the tradition over the years has been 
that universities provided the facilities for research. That 
came out of their tuition funds. It came out of their gifts 
from alumni and so forth, and the buildings were considered the 
university's responsibility. The money for research initially 
came primarily from donors, largely industry such as AT&T, and 
later on the universities took it upon themselves to also fund 
some of the research. Nowadays, universities are in tough 
financial times and so more and more they are trying to find 
other sources of money to build their buildings and their 
laboratories. The Federal Government has been fairly generous 
in funding equipment. That has just carried over. That has been 
considered a part of the responsibility of the central Federal 
Government. State governments contribute some as well but not 
as much. But it has always been considered the universities' 
requirement that they provide the buildings whether they get it 
from contributions or from state government, and that it how it 
has been for quite a while.
    What we are talking about today is what is the--and I am 
not trying to take words out of your mind or rephrase your 
objectives, Mr. Chairman, but basically the issue is, should 
the Federal Government be more active in providing not just 
grants for researchers to pay them or the graduate students or 
to buy equipment but should the Federal Government also become 
involved in building laboratories outside of the national 
laboratories, and that is a very important issue. I have mixed 
feelings about it. It would be a change in direction, although 
NSF has, by funding its Centers of Excellence, has gotten to a 
certain extent into the business of paying for buildings, but 
it is a very small part of their budget.
    So the issue we are discussing today, Mr. Chairman, I see 
as a very important issue and it is good for you to bring it 
up, but at the same time recognize that we are plowing new 
fields at this point by saying well, maybe the Federal 
Government should be providing more money for facilities, not 
just for salaries, not just for equipment but for the 
facilities that house these devices. And so I just wanted to 
emphasize that long history because as I say, it is a very 
important issue and we have to understand the history of it.
    In my formal statement, I go on to point out that 
universities have to have state-of-the-art science and 
technology facilities in order to remain at the cutting edge of 
research. With proper laboratory buildings, equipment and 
computing systems, students will graduate well prepared for the 
workforce. As the institutions represented here today know 
well, the most innovative students are attracted to 
universities with the best facilities. However, it is also 
important that we recognize that the National Science 
Foundation's expertise lies in the support of peer-reviewed 
basic research. They appoint panels of like-minded scientists 
to review proposals. These panels take it very, very seriously 
and they know that from personal experience because my son is 
currently sitting on the panel of the National Science 
Foundation and every few months he spends four days here in 
Washington, before that reads voluminous numbers of proposals 
and takes it very seriously and his colleagues on the panel do 
the same. It is not clear that the National Science Foundation 
has the expertise nor could it appoint the panels that have the 
expertise to research and judge proposals involving specialized 
facilities. Those new panels would have to be created, where we 
would get the people power to review it. I am just simply 
pointing out it would be quite a change for the National 
Science Foundation as they get very heavily into providing 
funding for facilities, buildings and so forth.
    We also recognize that science funding has changed 
dramatically in another area, and I apologize for going too 
long but I want to bring all these things together. When I was 
first elected to the Congress, Newt Gingrich asked me to write 
a science policy study to recommend science policy for the next 
generation. The last policy was written in 1945 by Vannevar 
Bush, and Newt and I both agreed that was a bit outdated, from 
1945 to 1995, and so I pointed out that with the increasing 
expense of major facilities, the Federal Government was no 
longer even going to be able to fund the research facilities, 
not just the buildings but also the equipment, because of the 
huge cost, and recommended that we should concentrate on 
international cooperation to do that, and that is come to an 
ITER, which the idea was originated in the United States. When 
we dropped it, the Japanese thought they would pick it up, then 
decided not to. Now the French are doing it with international 
collaboration, and that may be the direction of the future for 
the really expensive things such as CERN. CERN is another 
cooperative international effort which we are now joining. And 
it is very interesting to watch this evolution.
    As the projects get bigger and bigger, get more and more 
expensive, the buildings get extremely more expensive and so 
you are on the forefront of a major issue, Mr. Chairman, in 
terms of outlining where is the money going to come from and 
whether it is NSF or DARPA or ARPA-E or whatever. I think it is 
a major problem that has to be faced and discussed by the 
Congress and not just this Committee. So this gives us all a 
chance to do that.
    With that, I yield back.
    [The prepared statement of Mr. Ehlers follows:]

         Prepared Statement of Representative Vernon J. Ehlers

    Universities must ensure they have state of the art science and 
technology facilities in order to remain at the cutting edge of 
research. With proper laboratory buildings, equipment, and computing 
systems, students will graduate well-prepared for the workforce. As the 
institutions represented here today know, the most innovative students 
are attracted to universities with the best facilities.
    However, it is important that we recognize that the National 
Science Foundation's expertise lies in the support of peer-reviewed, 
basic research. That research often requires the use of various types 
of equipment and specialized facilities. Many scientific questions we 
are faced with today will only be answered through the use of very 
expensive facilities, such as ITER, that often require the 
participation of multiple countries to construct. There are also many 
smaller facility needs at our research institutions, some of which the 
NSF currently funds. However, I have some reservations about expanding 
this type of support, because it does not fit well into the primary 
mission and expertise of the NSF.
    I thank the Chairman for holding this hearing today, and look 
forward to hearing from our witnesses how we can support high-quality, 
sustainable research and infrastructure at our universities.

    Chairman Lipinski. Thank you, Dr. Ehlers. You knew after I 
said all those wonderful things about you that I wouldn't gavel 
you. I thought maybe you were going to filibuster until the end 
of your term.
    Mr. Ehlers. Well, that is a very good deductive process.
    Chairman Lipinski. I am an engineer, so--as you were 
telling that first story about Faraday and Ohm, I was wondering 
what the punchline was going to be, and I thought it might be 
something about Ohm's second law or something like that.
    With that, we are here to hear from the witnesses so maybe 
we should move on. We will have more time for this in the Q&A. 
If there are Members who wish to submit additional opening 
statements, your statements will be added to the record at this 
point.
    [The prepared statement of Ms. Johnson follows:]

       Prepared Statement of Representative Eddie Bernice Johnson

    Thank you, Mr. Chairman and Ranking Member for holding today's 
hearing on, ``The State of Research Infrastructure at U.S. 
Universities''.
    The United States' world-class university infrastructure is one of 
our greatest strengths that we must sustain to remain competitive in 
the 21st century. It is a bastion for academic research and education 
and a catalyst for scientific innovation. I am pleased that significant 
funding in the budget and the Recovery Act has been set aside towards 
expanding and sustaining our research and infrastructure.
    I am pleased that in 2010 the majority of our Nation's colleges and 
universities have access to high-speed internet. Additionally, 
according to the latest National Science Foundation (NSF) survey on the 
status of research facilities, 77 percent of universities nationwide 
rated their condition as satisfactory or superior. We must strive for 
excellence by funding initiatives and seeking new ways to improve our 
universities research facilities. Programs such as the Major Research 
Instrumentation (MRI) Program and the (ARI-R2) Program among 
others are good beginnings.
    As you are aware, nationally, our rate of increase in research 
space construction has slowed from 11 percent in 2003 to 3.7 today. 
These past few years, many colleges and universities have been hit hard 
by our recent economic recession, particularly Historically Black 
Colleges and Universities (HBCU's). Last year, $85 million in the 
additional funding was omitted from the federal budget for HBCU's. Many 
struggling universities have been forced to cut faculty and staff, 
delay construction, and furlough payrolls in order to survive during 
our current economic climate. Increased federal investment in our 
nation's research infrastructure is necessary to keep our colleges and 
universities competitive during these tough times.
    When considering the state of our national research infrastructure 
at U.S. universities we must protect those hurting the most first. For 
example, Historically Black Colleges and Universities graduate students 
in STEM degrees higher than most traditional universities and currently 
are conducting world-class research in AIDS and Cancer research. We 
must do what we can to help and protect minority serving institutions.
    While the topic of today's hearing is focused primarily on the 
research infrastructure of Universities, I must emphasize the bigger 
issue at hand which is increasing the number of U.S. students who 
enroll and graduate from these institutions. What sense does it make to 
have a world-class facility that is only half full? The United States 
faces a looming shortage of students enrolling in STEM disciplines. The 
key problem facing us right now is that we need more students in our 
university research infrastructure.
    I am interested in hearing from today's witnesses who are experts 
from universities across the nation. I thank you for your thoughtful 
testimonies as we look to address these issues. I am interested in 
hearing from you on how the federal government can strengthen our 
national university research infrastructure. I am also curious as to 
how our resources can be effectively distributed to institutions of 
higher learning that are in need the most.
    Thank you Mr. Chairman, I yield back.

    So at this time I am going to introduce our witnesses. 
First we have Dr. Leslie Tolbert, who is the Vice President for 
Research, Graduate Studies and Economic Development at the 
University of Arizona. We have Mr. Albert Horvath, who is the 
Senior Vice President for Finance and Business at Pennsylvania 
State University, as well as the Chair of the Board of 
Directors for the Council on Government Relations. I will now 
yield to my distinguished colleague, Mr. Inglis, to introduce 
our third witness.
    Mr. Inglis. Well, and I thank you, Mr. Chairman. It is 
wonderful to have Dr. John Raymond here today, the Vice 
President for Academic Affairs and Provost at the Medical 
University of South Carolina, a practicing nephrologist and a 
physician with the Department of Veterans Affairs. He has been 
with the Medical University since 1996, lending his expertise 
in medicine and academia to one of the strongest and oldest 
schools of medicine in the South. Dr. Raymond is also a 
prolific researcher. He has published over 100 full-length 
manuscripts and has received over $25 million in competitive 
research grants from the National Institutes of Health, the 
Department of Veterans Affairs and various foundations. As 
Chair of South Carolina's EPSCoR Committee, he plays a major 
role in expanding South Carolina's scientific and technical 
research. In 2009, South Carolina received a $20 million award 
from the National Science Foundation to work on biofabrication 
of human tissues. It is the largest NSF grant in our state's 
history. It will bring together the work of diverse South 
Carolina institutions including Greenville Tech and Furman 
University. The Medical University of South Carolina is playing 
a very significant role in this work and I know that Dr. 
Raymond shares my excitement about the great step forward this 
grant presents for both medical science and the research engine 
in South Carolina.
    Thank you, Dr. Raymond, for all the work you have done to 
expand opportunities in South Carolina and thank you for 
sharing your expertise with us today, and thank you, Mr. 
Chairman, for the opportunity to introduce this fine South 
Carolinian.
    Chairman Lipinski. Thank you, Mr. Inglis.
    And finally we have Dr. Thom Dunning, who is the Director 
of the National Center for Supercomputing Applications at the 
University of Illinois at Urbana-Champaign.
    As our witnesses should know, you will each have five 
minutes for your spoken testimony. Your written testimony will 
be included in the record for the hearing. When you have all 
completed your spoken testimony, we will begin with questions. 
Each Member will have five minutes to question the panel
    So we will start here with Dr. Tolbert.

    STATEMENT OF DR. LESLIE P. TOLBERT, VICE PRESIDENT FOR 
RESEARCH, GRADUATE STUDIES AND ECONOMIC DEVELOPMENT, UNIVERSITY 
                           OF ARIZONA

    Dr. Tolbert. Chairman Lipinski, Ranking Member Ehlers and 
other distinguished members of the Subcommittee, thank you for 
the opportunity to speak with you today on the state of 
research infrastructure in our Nation's universities. I am 
Leslie Tolbert. As you heard, I am the Vice President for 
Research, Economic Development and Graduate Studies at the 
University of Arizona in Tucson, Arizona. I am here speaking on 
behalf of the University of Arizona and also the Association of 
American Universities and the Association of Public and Land 
Grant Universities.
    The astounding research accomplishments in our universities 
have been the backbone of our country's economic 
competitiveness, our high living standard and our national 
security for over 60 years. In recent decades, though, our 
global leadership position in innovation is being threatened.
    As the major provider of support for our university-based 
research, the Federal Government must act quickly to build on 
the American Recovery and Reinvestment Act of 2009 to make 
continued strategic investments in research itself, and also 
the subject of today's hearing, in the physical foundation for 
that research, which includes research instrumentation and 
cyberinfrastructure as well as bricks-and-mortar laboratories.
    In the current economic crisis, state support for research-
related expenses in public universities and charitable giving 
and returns on endowment to universities have declined 
precipitously. Federal support for research is as important as 
ever before. Federal support for research infrastructure comes 
in part through the provision of facilities and administration, 
or F&A, cost recovery that is included in grant awards. F&A is 
intended to reimburse universities for the collective hidden 
but real expenses of research. But growing federal mandates, 
for research compliance in particular, consume more and more of 
those funds, leaving little to support the physical 
infrastructure. The Federal Government also has embedded in a 
number of agencies mechanisms that are designed specifically to 
fund research infrastructure. I will mention some of these in a 
moment but first let me discuss the situation at my university, 
which I think provides a useful model for study.
    The University of Arizona, or UA, is a large comprehensive 
land grant university with annual R&D expenditures in science 
and engineering amounting to $550 million. We are one of the 
top 25 research universities in the Nation, and we are number 
one in the physical sciences in the latest NSF rankings. UA 
revenues come roughly one-quarter from state appropriations, 
one-quarter from student tuition and fees, and half from other 
sources, which includes sponsored grants and contracts, which 
come primarily from the NIH, the NSF and NASA.
    The State of Arizona appropriations for its three public 
universities have fallen steeply in the last two years, 
precipitating the deepest crisis in recent history. In these 
last two years, the UA's annual state appropriation has been 
reduced by $100 million. Founded in 1985, the UA has many 
science buildings that are old and deeply in need of repair. 
State funding for new buildings and for building maintenance 
has been very small over the years. For building maintenance 
and repair in the last decade, we received $11 million from the 
state to address a documented need that now exceeds over $200 
million.
    Given our constraints, careful planning is essential. Our 
current planning efforts are driven by several principles, two 
of which are using flexible open design laboratories. That is 
economical. They promote collaboration and they allow 
individual projects to grow and shrink without the expense of 
moving walls. And another thing we do is, we support shared 
equipment facilities over facilities that are dedicated to 
individual researchers, to maximize the impact of these 
facilities.
    It has become clear at the UA and across the country that 
more federal support is needed for the research infrastructure 
that enables federally funded research. My colleagues and I 
would recommend that the NSF strive to increase the percentage 
of the budget that it spends on infrastructure from 24 to 27 
percent by fiscal year 2016. Building to that level gradually 
as the overall NSF budget grows should minimize the impact on 
funds available for research grants.
    The increasing infrastructure support, I would argue, 
should go to several programs. One is the Academic Research 
Infrastructure program that provides critical funding to 
modernize existing research laboratories. Funding from ARRA 
this year was very welcome and we point to the importance of 
continuing that program beyond ARRA. The Major Research 
Instrumentation and Major Research Equipment and Facilities 
Construction programs are also essential to provide state-of-
the-art research equipment priced in the range of several 
millions of dollars or above tens of millions of dollars. In my 
written testimony, I have outlined several specific changes in 
these funding mechanisms that would give them greater impact.
    Perhaps most importantly, we recommend that OSTP create a 
national science and technology working group to assess 
instrumentation and infrastructure programs at all federal 
agencies, and make recommendations concerning steps that the 
government should take to ensure adequate support for its 
national research infrastructure. Our researchers, if armed 
with direct research funding and a strong research 
infrastructure, can continue to lead the world in innovation 
and discovery.
    Thank you for the opportunity to present this testimony.
    [The prepared statement of Dr. Tolbert follows:]

                Prepared Statement of Leslie P. Tolbert

    Chairman Lipinski, Ranking Member Ehlers, and other distinguished 
Members of the subcommittee, thank you for the opportunity to speak 
with you today on the state of research infrastructure at our nation's 
research universities.
    My name is Leslie Tolbert. I serve as the Vice-President for 
Research, Graduate Studies, and Economic Development at the University 
of Arizona, in Tucson, Arizona. I am honored to have the opportunity to 
offer testimony on behalf of the University of Arizona, the Association 
of American Universities, and the Association of Public and Land-grant 
Universities.

Overview

    Our nation's research universities are falling behind in their 
ability to provide the physical infrastructure--both the laboratory 
buildings and the high-end technical facilities in those buildings--
needed to keep our researchers working at full capacity. As state and 
private sources of funding dwindle, even more quickly during the 
current economic slump, federal support is growing in importance. 
Strategic investments in research infrastructure by the federal 
government are absolutely essential to maintaining a global leadership 
position for U.S. science. The University of Arizona, with its sharply 
declining support from the state, provides a useful model for 
understanding the current situation in a large public research 
university and the specific remedies that federal resources could 
provide.

Background and Context

Funding for University-based Research

    The record of research accomplishment of U.S. universities is 
astounding. For the past 60 years, at least, these accomplishments have 
been the backbone of our economic competitiveness, high living 
standard, and national security. As documented in the National 
Academies ``Rising Above the Gathering Storm'' report, our leadership 
position in education and innovation has been threatened in recent 
decades as other countries have sought to emulate us by making huge 
investments in their research enterprises. U.S. leadership in science 
and engineering will be maintained only if we maintain a modern and 
effective research infrastructure.
    For many decades, the federal government has assumed the 
responsibility of providing the dominant support for university-based 
research and research training, providing billions of dollars in 
support, virtually all of it on a competitive basis to ensure that the 
most meritorious research ideas receive funding. Our system of 
competition through review and ranking of applications by peers is the 
envy of the world.
    In recent years, however, federal support of university research in 
science and engineering, while still substantial, had become 
essentially flat in real dollars (AAAS Report XXXII Research and 
Development FY 2008, Chapter 2: Historical Trends in Federal R&D 
http://www.aaas.org/spp/rd/08pch2.htm), even while that of other 
countries was growing. The American Recovery and Reinvestment Act of 
2009 has provided much needed federal funding to reverse this trend for 
two years, but it is unclear what the picture of federal research 
support will look like after ARRA funding ends.
    Adding to the problem, as the states have faced growing economic 
challenges, state support for research-related expenses in many public 
universities has declined precipitously, and charitable giving and 
endowment returns to both public and private institutions have also 
fallen sharply (Council for Aid to Education, http://www.cae.org/
content/pdf/
VSE-2009-Press-Relsease.pdf, and 
National Association of College and University Business Officers, 
http://www.nacubo.org/Documents/research/
2009-NCSE-Press-Release.pdf). As a 
result, American research productivity and scientific advances are 
likely to diminish. The private sector spends more than twice as much 
as the federal government spends on research and development (National 
Science Board, Science and Engineering Indicators 2010), but in tight 
economic times, private industry is driven increasingly to focus its 
research dollars on applied research and development for short-term 
profit, leaving to the universities the basic research--and unexpected 
discoveries--that ultimately must form the basis for future 
applications.
    Maintaining America's universities' competitiveness in fundamental 
research and research-enriched education has become a serious 
challenge. Meeting this challenge will require strategic investments in 
the physical infrastructure for research as well as in the research and 
educational activities themselves.

Funding for Physical Infrastructure for Research in Universities

    The physical infrastructure for research includes not just bricks-
and-mortar buildings, but also research instrumentation and a robust 
cyberstructure (for internal and external communication and for 
research requiring high-performance computing). The increasing 
complexity of science and engineering requires advanced technical 
equipment and tools, as well as specialized workspaces that encourage 
and enhance collaboration and interdisciplinary pollination of ideas.
    The physical infrastructure that must be in place for cutting-edge 
research was historically provided by a combination of federal and 
state government and university funds. Federal dollars for 
infrastructure have decreased, however. As described by Homer Neal, 
Tobin Smith, and Jennifer McCormick in their book, Beyond Sputnik--U S. 
Science Policy in the 21st Century (U. Mich. Press, 2008):

         ``In the years following World War II and immediately after 
        Sputnik the .US government invested heavily in the development 
        and funding of scientific infrastructure at universities, 
        national laboratories, and other federal research facilities. 
        However, by the early 1970's many federal programs that had 
        previously existed to support construction and renovation of 
        research facilities ended, and federal obligations for research 
        facilities and large equipment in colleges and universities 
        dropped significantly. During this period, the neglect of 
        laboratory instrumentation and the erosion of the physical 
        infrastructure for research threatened the long term vitality 
        of even leading universities.''

    Today, federal dollars are directed primarily to supporting 
research operations, with little targeted directly to the costs of 
providing the necessary research infrastructure. To fill this gap, 
universities have relied heavily upon state support, endowments, gifts, 
and other institutional resources to support their physical research 
infrastructure needs. However, declines in state support for public 
universities and in endowments and gifts for public and private 
universities, have made it increasingly difficult for us to sustain and 
renovate existing laboratories or to build the new facilities that are 
required for increasingly sophisticated research.
    As a result, universities are falling behind the need to provide 
the physical facilities to do the research that will propel our economy 
forward. According to the National Science Board's 2010 report of 
``Science and Engineering Indicators:''

         ``Research performing colleges and universities continued a 
        two-decade trend of increasing the amount of research space at 
        their institutions. [. . .] In recent years though, the rate of 
        increase in research space has begun to slow. [. . .] The rate 
        of increase peaked in FY 2001-03 at 11%. Since then, the rate 
        of increase has gradually declined [. . .] In conjunction with 
        the slowdown in the increase in research space, the total 
        amount of newly constructed research space also began to slow 
        at the beginning of the decade (table 5-5). Since FY 2002-03, 
        the total amount of new research space constructed declined by 
        approximately 45%.''

Current Situation Regarding Federal Support for Research Infrastructure

    The federal government provides support for research infrastructure 
in several ways. Some support for research facilities comes through the 
provision of Facilities and Administration (F&A), or ``indirect,'' cost 
recovery that is included in grants and contracts awards. F&A cost 
recovery is intended to reimburse universities for expenditures on the 
buildings, utilities, equipment, libraries, and administration that 
collectively support their research.
    A large portion of the funds awarded for F&A costs are, in fact, 
not available for the kinds of infrastructure projects I have 
mentioned. Most notably, growing federal mandates and research 
compliance requirements have pulled institutional funds away from 
support of research facilities. A 2004 report from the Council on 
Government Relations (``A New Research Business Model: Incentivizing 
Research'') points out that universities actually provide significant 
cost-sharing:

         ``Universities contribute to the direct costs and the indirect 
        (i.e., F&A) costs of federal research. The National Science 
        Foundation's (NSF) annual survey on Research and Development 
        (R&D) Expenditures at Universities and Colleges shows the 
        significant and increasing financial contributions made by all 
        colleges and universities, in total, to the research enterprise 
        over the past sixty years. [. . .] when shown as a percentage, 
        the important role of Institutional Funds is clear. Over the 
        period from 1976 to 2006, the share of R&D expenditures in this 
        category has grown faster than any other category. According to 
        the 2006 NSF Survey, Institutional Funds account for 19.0% of 
        all R&D expenditures, compared to 12.0% of all R&D expenditures 
        in 1976. To put this in another context, the increased share 
        from 12.0% to 19.0% represents a growth factor of 58%.''

    In addition, there are a limited number of federal mechanisms 
designed specifically to fund research infrastructure. These include 
NSF's Major Research Equipment and Facilities Construction (MREFC) 
program and their Major Research Instrumentation (MRI) program; NIH's 
Shared Instrumentation Grants and High-End Instrumentation Grants; the 
NCRR Animal Facility, Research Facility Improvement (C06), and Core 
Facility Renovation, Repair, and Improvement (G20) programs; research 
facility construction funds from the National Institute of Standards 
and Technology; and the Department of Defense's University Research 
Instrumentation Program. Some of these infrastructure programs and 
their scopes were temporarily expanded with the use of American 
Reinvestment and Recovery Act (ARRA) funds. One program that was 
revived with ARRA funds was the NSF's Academic Research Infrastructure 
(ART) program, which I will discuss further in my recommendations.

How the University of Arizona Supports Research Infrastructure

    At the University of Arizona, one can see firsthand the impact of 
all the aforementioned issues, including the precipitous decline in 
state funding as well as the shrinking funding for research 
infrastructure from federal sources. I think you will find the UA to be 
a useful case study.
    The University of Arizona is a large, comprehensive land-grant 
university that includes, together on one campus, liberal arts colleges 
and colleges of medicine, pharmacy, nursing, public health, 
engineering, optical sciences, and law. On a separate campus, we have a 
Science and Technology Research Park. We are one of the top 25 research 
universities in the nation and a member of the Association of American 
Universities. In FY 2008, our Science and Engineering Research and 
Development expenditures amounted to $546 million; we were ranked #1 in 
total R&D expenditures in physical sciences by the NSF. Approximately 
27% of our operating expenses are in support of research.
    In FY 2010, 22% of UA revenues were from state-appropriated funds; 
27% were from student tuition and fees; and the remaining 51% were from 
other sources, including sponsored grants and contracts, auxiliary 
funds, gifts, and investment income. [See Table 1 below.] Each year, 
sponsored grants and contracts come primarily from the federal 
government, with the remainder from industrial sponsors, foundations, 
and private contributions. Among federal sponsors, the Department of 
Health and Human Services (HHS) provides the largest single share of 
sponsored grants and contracts (primarily via the National Institutes 
of Health), followed by NASA, National Science Foundation, Department 
of Defense, and Department of Agriculture.



    To date, we have been awarded $83.7 million (including anticipated 
year 2 amounts) in ARRA federal stimulus funds for a wide range of 
important projects on topics ranging from solar electric materials to 
optical imaging methods for cancer detection to methods for monitoring 
soil moisture in arid lands. Most of the ARRA support is for research 
projects; $4.7 million from the U.S. Department of Commerce supports a 
new biotechnology park; and just under $1 million from NSF is for 
research equipment.
    Another federal funding source from which we will receive support 
in the near future is the MREFC program at NSF. We will serve as the 
Southwest's core site for the National Ecological Observatory Network, 
or NEON, for regional--to continental-scale ecological research. The 
project has recently passed its Final Design Review and the President's 
FY 2011 budget proposes $433M in MREFC funds to begin the construction 
phase of NEON. The exact amount of funding that will flow to the UA is 
not yet determined.
    In contrast to federal funding, State of Arizona support of its 
public universities has fallen steeply in the last two years, 
precipitating a crisis deeper than any other in recent history. As 
shown above in Table 1, the percent of the UA budget that comes from 
the state has fallen from 34% to 22% in the past decade. Table 2 below 
shows the dramatic decline in just the last three years, from $443 
million appropriated to the UA in FY 2008 to $348 million appropriated 
in FY 2010.



    Our research buildings range from modern and well-equipped to 
outdated and deeply in need of maintenance. The university was founded 
in 1885, and most of our science related buildings were built in the 
1960s through the 1990s. Our older buildings do not meet current safety 
codes, limiting their utility for research involving hazardous 
biological or chemical agents. With their small, compartmentalized 
spaces, they certainly are not conducive to current modes of 
collaborative research. We struggle to find the resources to update 
those buildings, as well as to build new research buildings that can 
provide the new lab space that we need.
    We received no State of Arizona funds for new building projects 
between the early 1980's and FY 2008. House Bill 2529, signed into law 
in 2003, provided significant relief in the form of state 
appropriations of over $440 million for Research Infrastructure 
Financing for the three state universities over 23 years (FY 2008--FY 
2031). From HB 2529, the UA receives $14 million per year for debt 
financing. Table 3, below, shows the sources of funding for our ten 
most recently constructed research buildings.



    A major shortage of state support for Building Renewal at the 
universities contributes to the challenges of using existing aging 
buildings for research. The state has a formula forcalculating Building 
Renewal needs based on the replacement values and ages of our 
buildings. The state provided only partial funding for the 
universities' Building Renewal needs in 1987-2001, and has failed to 
provide any Building Renewal funds for eight of the past nine years. 
Over the past five years, FY 2006-10, we should have received $200 
million. Instead, we received only $10.9 million, in FY 2007, thus 
falling short by $178 million for this period ,alone. Added to the 
shortfalls from before 2001, this leaves the UA with an accumulating 
Building Renewal need that far exceeds $200 million in FY 2010. Old 
chemistry and engineering buildings are in particular disrepair and can 
not be used for most types of research in their nominal fields.
    In sum, it has become clear that the state cannot fund the 
improvements needed to keep pace with emerging research needs, and the 
university struggles to fund the improvements needed to comply with 
general laboratory safety codes as well as emerging research needs. To 
guide that struggling effort, the UA has a Space Committee, chaired by 
the Provost and the Senior Vice President for Business Affairs. The 
Committee plans building renewal and construction, assessing and 
balancing research infrastructure needs against the availability of 
funding and a university-wide commitment to safety and environmental 
stewardship.
    Our conceptual framework for efficient, cost-effective campus 
build-out addresses several key issues:

          We have accepted an urgent mandate to protect the 
        environment even as we continue to build. When addressing space 
        needs, we first consider refurbishment of old buildings. Often 
        it is too expensive to upgrade existing research facilities, so 
        older research space is converted to offices and classrooms 
        instead. When new buildings are needed, we are committed to 
        building them to at least LEED silver specifications, which is 
        more expensive in the short run but will provide future energy 
        savings to help offset the expense. For laboratory research 
        buildings, which use more energy than office buildings, these 
        savings over time can be great.

          We build out the campus utility infrastructure sector 
        by sector, rather than building by building, in accordance with 
        our campus master plan and capital improvement planfor the 
        coming 5-10 years. This coordinated approach is very 
        economical, allowing the infrastructure and new buildings to be 
        constructed as efficiently and inexpensively as possible. For 
        example, we have applied for a $15M NIH C06 ARRA grant to build 
        a new research building for imaging sciences. The building 
        construction cost and schedule are greatly reduced because 
        utility infrastructure is already in place. Thus, any funding 
        received will be most effectively used for its core research-
        support purposes.

          New laboratory buildings generally have a flexible 
        open-laboratory design. This is economical, promotes 
        collaboration among research groups, and allows space for 
        particular projects to grow and shrink as funding waxes and 
        wanes, without the expense of moving walls or utility spines. 
        This approach leads to research funding and discovery successes 
        that would otherwise not occur. Within a few years of the 
        opening of our new open-configuration, interdisciplinary life 
        sciences building, our faculty landed a $50M NSF grant (the 
        largest ever to an Arizona institution) to support 
        collaboration of molecular plant biologists, ecosystems 
        biologists, information scientists, earth-imaging specialists, 
        and others to tackle Grand Challenge problems in plant biology.

          Shared equipment facilities are preferable to 
        facilities under the control of individual researchers. At the 
        centers of our new open-lab buildings are shared core 
        facilities for the most expensive instruments they need, such 
        as those for microscopy, genomic and proteomic analysis, and 
        high-end computing. These core facilities are an economical way 
        to provide large numbers of researchers access to the latest 
        equipment, equipment that they could not afford on their 
        individual grants.

    The UA has built ten new research buildings in the past ten years 
and our Capital Plan includes plans to build three more in the coming 
two years, to meet our most urgent projected needs. One of these, a 
research support building for our new College of Medicine arm in 
Phoenix (in partnership with Arizona State University), will be funded 
primarily with ARRA funding through an NIH C06 award. Incidentally, the 
development of that entire medical campus has been a collaboration of 
many entities dedicated to research advancement, including the UA, the 
City of Phoenix, and public-private partnerships.
    Our recently constructed buildings, in both Tucson and Phoenix, are 
funded by a combination of state and local funds. Projected sources of 
funds for the next three new research buildings and for research-
related renovations on our Capital Plan are shown in Table 4, below. We 
take advantage of the State of Arizona's recently approved Stimulus 
Plan for Economic and Educational Development (SPEED), a creative 
mechanism whereby the State will provide 80% of annual debt service 
payment from state lottery funds, while the universities will cover 20% 
of the annual debt service payments through institutional funds (which 
include student retained collections; State appropriations; and 
indirect cost recovery). Indirect cost recovery alone will be expected 
to cover approximately 10% of the debt service.



    In addition to building renewal and construction, we track our 
expenditures on capital equipment (item cost >$5,000). While the total 
investment in capital equipment varies year to year, the percent 
contribution from federal funds has declined systematically in recent 
years, from 68% in FY 2003 to just 46% in FY 2009. [See Table 5 below.] 
Thirteen percent ($10 million) of our equipment purchased with federal 
funds in the past ten years has been purchased with funds designated 
for shared-use instrumentation.



    In addition to our primary campus in Tucson and second medical 
campus in Phoenix, we have a Science and Technology Park in the 
outskirts of Tucson. With more than 7,000 employees, the UA Tech Park 
reflects one aspect of our partnership with the private sector in 
regional development and is one of the region's largest employers. It 
is home to 40 high-tech companies and business organizations, including 
several emerging technology companies, as well as branches of five 
Fortune 500 companies. It includes a business incubator, which 
currently hosts 12 emerging companies, several of which are spin-offs 
from the university. The Park is an independent legal entity [501(c)3]. 
We currently are developing a second UA Tech Park, focused on 
biotechnology, closer to the UA campus, and recently received $4.7 
million in ARRA funds from the U.S. Department of Commerce to build the 
utility and roadway infrastructure that will allow us to develop the 
property.

Gaps in Our Ability to Provide Necessary Research Infrastructure

    All of the innovative collaborations and approaches being used to 
facilitate leading-edge research require new or upgraded research 
facilities, for which there is currently insufficient funding. Under 
current conditions, many of these needs will likely go unmet.
    As we seek multiple funding sources and new arrangements to fund 
building renewal and upgrades, the UA and other universities across the 
country face a specific and severely hobbling gap in funding 
opportunities. Donors may be willing to help to fund new buildings, but 
they are very rarely willing to contribute to ongoing operations, 
maintenance, or upgrades. For lack of funds, maintenance and upgrading 
are often deferred or neglected. Allowing our universities' older 
research buildings to languish raises the future costs of providing the 
necessary physical research infrastructure. As discussed earlier, the 
University of Arizona has a growing need for refurbishment of its 
buildings that exceeds $200 million today.
    Beyond a shortage of funds for building renewal, universities face 
a confounding problem: a gap in funding opportunities for mid-scale 
instrumentation facilities. NSF's Major Research Equipment and 
Facilities Construction (MREFC) program supports the acquisition, 
construction, and commissioning of large scale research facilities and 
equipment, in the tens to hundreds of millions of dollars range, that 
uniquely advance the frontiers of science and engineering. Initial 
planning and design, as well as follow-on operations and maintenance 
costs of the facilities, are provided. NSF's Major Research 
Instrumentation (MRI) program funds the acquisition or development of 
single pieces of research instrumentation, up to $4 million in cost (or 
$6 million, with ARRA funds), that are to be shared by multiple 
investigators. The program explicitly does not support acquisition or 
development of the whole suite of instruments that is often needed to 
outfit high-end research facilities. Similarly, the NIH has a Shared 
Instrumentation Grant (SIG) program that supports the purchase of 
instruments up to $600,000 in cost. The huge gap between these two 
funding mechanisms and the MREFC makes it very difficult to fund 
medium-scale infrastructure.
    A smaller but still constraining issue arises from the fact that 
the MRI and SIG programs support the purchase or development of 
expensive pieces of scientific instrumentation, but do not provide for 
the renovations that often are needed for installation of the new 
instruments and do not provide for personnel, ancillary equipment, and 
upgrades to keep the instruments at the cutting edge as technology 
advances. In addition, the MRI program requires universities to provide 
30% in matching dollars. Because of the difficulty in finding the funds 
to fulfill those particular requirements, we are sometimes unable to 
apply for needed instruments, and even if we do obtain the funds to 
purchase new items, good instrumentation may fall away from the cutting 
edge, even when relatively inexpensive upgrades could have kept them up 
to date.

Recommendations

    In light of the severity of the issues I have raised, we recommend 
the following:

        1)  The NSF should increase the percentage of its budget that 
        it spends on infrastructure to 27 percent by FY2016--in 
        accordance with the recommendation made by the National Science 
        Board in its 2003 report, ``Science and Engineering 
        Infrastructure for the 21'' Century: the Role of the National 
        Science Foundation,'' (http://www.nsf.gov/od/lpa/news/03/
        pr0340.htm).

    Recent figures suggest that NSF currently devotes some 24 percent 
of its funding to infrastructure support. As the Congress and the 
Administration seek to double funding for the agency by FY 2016, we 
believe the 27 percent target set forth by the National Science Board 
is a reasonable goal. Moreover, slowly increasing the percentage of 
funding NSF devotes to infrastructure over five years as the overall 
NSF budget grows should minimize the negative impact on the funds 
potentially available for research grants and awards.
    To help to achieve this goal, we specifically recommend that:

        a.  The Congress and NSF should continue to support the Major 
        Research Instrumentation (MRI) and Major Research Equipment and 
        Facilities Construction (MREFC) programs.

    These programs are essential to provide state-of-the-art research 
equipment priced in the range of several millions of dollars or above 
tens of millions of dollars. It would be especially helpful for MRI 
grants in the future (1) to fund not only the purchase of the 
equipment, but also renovations, ancillary equipment, and personnel 
that may be needed to put those instruments to best use, and (2) not to 
require the significant (30%) matching dollars currently required of 
universities. Absent that additional support, the full costs of 
providing new technical capabilities are so high that some universities 
are unable to participate in the MRI program.

        b.  The Committee should authorize and funds should be 
        appropriated for the Academic Research Infrastructure (ARIA 
        program to enable NSF to solicit proposals to make additional 
        ARI awards beginning in FY 2012.

    Renovation of existing facilities is a critical need for which it 
is often difficult to find funding solutions. The inability to 
modernize existing research facilities often decreases research 
productivity, meaning that the value of the research funding provided 
is not fully leveraged, as researchers are forced to conduct their 
research in suboptimal facilities.
    The NSF Academic Research Infrastructure (ARI) program was 
originally authorized to try to address this very issue. Unfortunately, 
the program was never very well funded and its last solicitation was in 
1996 which is, in part, why the funding provided with ARRA dollars this 
year for the ARI-R2 program was received so favorably by the 
universities I represent here today. The program is right on the mark, 
aimed at modernizing existing shared research facilities. It will be 
important in helping to ensure that our research infrastructure keeps 
pace with our science--that is that the research that NSF funds can be 
done in appropriate research facilities--but it is funded for one year 
only, at $200 million. In its single solicitation, it received 
proposals for $1.02 billion in projects. Extension and expansion of the 
ARI program, through authorization and funding in FY 2012 and beyond, 
is critical, and the return on this investment will be high. Placing 
the emphasis on shared facilities ensures maximum impact per dollar.

        c.  The NSF should develop a new program to support mid-scale 
        infrastructure projects not currently eligible for support 
        through the MRI and MREFC accounts.

    Such a program would be a significant means to support major 
research infrastructure needs. The National Science Board (NSB) has 
identified several specific areas where mid-scale infrastructure is 
needed. These areas include: acquisition of an incoherent scatter radar 
to fill critical atmospheric science observational gaps; replacement or 
upgrade of submersibles; beam line instrumentation for neutron science; 
and major upgrades of computational capability.
    As the 2003 NSB report on scientific and engineering infrastructure 
noted, ``In many cases the midsize instruments that are needed to 
advance an important scientific project are research projects in their 
own right, projects that advance the state-of-the-art or that invent 
completely new instruments.'' Thus, this program would advance the 
state of research technology, as well as spread the use of such high-
end technologies.

        2)  OSTP should convene a National Science and Technology 
        working group to assess the effects of the serious decline in 
        state and private funding for university research 
        infrastructure and recommend steps by the federal government to 
        ensure adequate support for the nation's academic research 
        infrastructure.

    The need for such analysis and thought on the financial future of 
research universities is so dire that, in multiple forums, university 
leaders across the country already are convening for discussion of, 
among related topics, specific research infrastructure needs and the 
most effective solutions that could be implemented. An OSTP working 
group could incorporate the perspectives of individual agencies and 
these university discussions to move the national conversation forward 
with focus, in time for deliberations around the 2012 budget 
formulation.
    Specifically the OSTP working group should:

        a.  Assess existing and propose new research instrumentation 
        and infrastructure programs at all federal agencies, including 
        those recommended above for the NSF.

    In recent years, the funds available for research infrastructure 
programs outside of NSF, such as those supported by the NIH's National 
Center for Research Resources Division of Research Infrastructure, have 
dwindled. Meanwhile, the need and demand for these programs remains 
very high. As just one example, NIST's competitive university 
facilities construction grant program, which received funding of only 
$24 million in FY08, was able to support only three out of 93 
proposals. Through additional funds provided to this program in FY 09 
and through ARRA, NIST has been able to go further to address some of 
the pent up demand for new research facilities, however, the demand is 
still very high. Moreover, this demand will only grow as we move to 
increase the amount invested in research activities at key agencies 
such as the NSF, Department of Energy Office of Science, and NIST, as 
called for by the President and in the America COMPETES legislation 
which this committee will be looking to reauthorize this year.

        b.  Conduct a critical review of the increasing financial 
        pressures that impede the ability of research universities and 
        other institutions to adequately support critical physical 
        research infrastructure needs.

    In recent years the amount that universities, including the 
University of Arizona, have had to spend to ensure compliance with an 
increasing array of federal regulations has dramatically increased, 
requiring a significant amount of university revenues to go to 
supporting a greatly expanded ``research compliance infrastructure.'' 
Many of these costs are not currently reimbursable by our sponsoring 
agencies. Thus, they must be paid out of the universities' own 
institutional funds, draining the resource pool that otherwise is 
available to help to support the university's physical infrastructure 
needs. The increasing financial pressure, as well as the impact of 
increasing cost sharing requirements on universities, should be 
carefully examined.

Conclusion

    The National Academies ``Rising Above the Gathering Storm'' report 
had proposed that the government:

         ``Institute a National Coordination Office for Advanced 
        Research Instrumentation and Facilities to manage a fund of 
        $500 million in incremental funds per year over the next five 
        years--through reallocation of existing funds or, if necessary, 
        through the investment of new funds--to ensure that 
        universities and government laboratories create and maintain 
        the facilities, instrumentation, and equipment needed for 
        leading-edge scientific discovery and technological 
        development. Universities and national laboratories would 
        compete annually for these funds.''

    While we stop short of endorsing the specific amount of funding for 
infrastructure programs across all government agencies, we feel that 
there clearly is a need for a revitalization of existing agency 
infrastructure programs as well as the development of new programs. It 
is therefore time that the Congress, OSTP, and all federal agencies 
work together to conduct a serious assessment of what the government 
can do to ensure that research infrastructure needs required to support 
government-sponsored research activities are being met adequately.
    The significant amount of money devoted to research infrastructure 
programs in ARRA provided a critical shot in the arm which helped to 
inoculate the nation against the effects of years of neglect of our 
research infrastructure. That being said, additional federal support 
for research infrastructure is still very much needed after ARRA funds 
end, to carry forward our ability to meet the significant needs that 
exist for renovation and upgrade of aging facilities across the 
country. This is particularly true in light of declining alternative 
funding sources that universities have traditionally been able to rely 
upon to support their infrastructure needs.
    The return on this investment will be high. Our researchers, armed 
with direct research funding and supported by a strong research 
infrastructure, will be able to continue to lead the world in 
innovation and discovery. At my own institution, we have seen what can 
happen when modem infrastructure is made available: our faculty members 
almost certainly would not have landed the $50 million grant from the 
NSF to address major global issues in plant biology if they had not 
been located in well-outfitted facilities that were designed to enhance 
cross-disciplinary collaboration.
    Thank you for the opportunity to present this testimony.

                    Biography for Leslie P. Tolbert

    Leslie P. Tolbert, Ph.D., has been the Vice President for Research, 
Graduate Studies, and Economic Development of the University of Arizona 
since 2005. In this role, she supports the research and other scholarly 
activities of a $565M research enterprise, promotes technology transfer 
and commercialization, and oversees the graduate programs of the 
university. Dr. Tolbert received her undergraduate and Ph.D. degrees 
from Harvard University and obtained postdoctoral training at Harvard 
Medical School. She has been on the neuroscience faculty of the 
University of Arizona since 1987. She currently is a Regents' Professor 
with appointments in the College of Science and in the College of 
Medicine. For over 25 years, she has led a research group that 
investigates the critical role of sensory input in guiding the 
development of the brain. Dr. Tolbert has served as president of the 
Association of Neuroscience Departments and Programs and of the 
Association for Chemoreception Sciences, is active in the Society for 
Neuroscience and the American Association for the Advancement of 
Science, and is a member of numerous boards, including those of the 
Arizona Center for Innovation, the Arizona Alzheimer's Consortium, and 
the Large Binocular Telescope. She has served on numerous review and 
advisory panels of the National Science Foundation and National 
Institutes of Health.

    Chairman Lipinski. Thank you, Dr. Tolbert.
    Mr. Horvath.

 STATEMENT OF MR. ALBERT G. HORVATH, SENIOR VICE PRESIDENT FOR 
FINANCE AND BUSINESS, PENNSYLVANIA STATE UNIVERSITY, AND CHAIR, 
      BOARD OF DIRECTORS, COUNSEL ON GOVERNMENT RELATIONS

    Mr. Horvath. Thank you, Mr. Chairman and members of the 
Subcommittee for allowing me to participate as part of this 
distinguished panel.
    Research is central to the mission at major universities 
like Penn State. The research enterprise is complex, requiring 
a robust infrastructure that is modern and flexible. Elements 
include dedicated buildings, specialized equipment, high-end 
computing and substantial administrative support. Such an 
infrastructure is complicated and expensive to develop and 
maintain. Universities have had to subsidize the physical and 
administrative infrastructures supporting research with 
revenues other than funding from sponsors. Major infrastructure 
investments are commonly financed with substantial amounts of 
long-term debt, based upon an expectation that research funding 
will continue over the term of the issued bonds. The bonds are 
repaid used unrestricted revenue sources, including a portion 
of recovered facility and administrative costs, F&A, on 
federally sponsored research awards.
    Three recent events have caused anxiety around the 
sustainability of these expensive assets. First, unpredictable 
funding available for major federal research sponsors puts 
pressure on universities' plans to service its external debt. 
Second, the economic downturn of the last 18 months has reduced 
endowment values and constrained future borrowing potential. 
And third, other sources of possible funding for research 
facilities, namely philanthropy and state investments, have 
fallen considerably because of the economic downturn. These 
events have dampened the ability of research universities to 
invest in all facilities, including those supporting research.
    In light of this challenging environment, the most 
important element toward encouraging investment in the research 
infrastructure is a reliable stream of direct research funding. 
If a long-term commitment is made to predictable growth in 
funding agencies' budgets, it would help to provide the 
confidence that revenues would be available to support the 
activities within them over a 20- to 30-year period. We 
strongly encourage the continuation of programs that have 
helped to support research infrastructures such as the ARI and 
other similar programs. They have provided critical assistance 
in our ability to meet the needs of our faculty and research 
staff.
    Consider support for research-focused capital investments 
designed to help reduce the risk of these long-term financial 
commitments. This could be in the form of low-interest capital 
made available to universities for specific types of projects, 
or rate subsidies for borrowings made directly by the 
university. The success of the Build America bonds program over 
the past year is an indication of how modest investments by the 
Federal Government confer significant economic activities.
    Allow universities to cover the costs of money on internal 
funds used to finance research infrastructure investments. This 
is an allowable cost for commercial recipients of federal 
funds. The change to our cost accounting guidelines can send 
more allocation of university reserves to such projects.
    We would urge consideration of the removal of the cap that 
was placed upon administrative cost recovery by universities 
almost 20 years ago. Most of these universities now exceed the 
26 percent cap, resulting in millions of dollars worth of 
legitimate research support expenses that go unreimbursed. No 
other type of contractor performing work for the Federal 
Government is subject to such a cap on supportable allocable 
costs.
    A new, uncapped pool for regulatory compliance costs should 
be considered if removal of the administrative cost cap is not 
deemed feasible. Since implementation of the cap, several new 
requirements and regulations have been enacted that require 
greater effort by universities. However, none of the 
incremental costs associated with the regulatory changes can be 
recovered.
    Require all federal sponsors to reimburse F&A costs at 
approved federal rates. A study by the RAND Corporation in 2000 
estimated that universities are subsidizing federally sponsored 
research by as much as $1.5 billion that would be eligible for 
reimbursement if all sponsors were paying the approved F&A 
rate. Dozens of new and expanded federal regulations on the 
conduct of research implemented since 2000 have surely added to 
the university subsidy.
    In conclusion, the research partnership between the Federal 
Government and American research universities has enabled great 
achievements in science, yielded innovation and economic growth 
and helped our system of higher education become the envy of 
the world. It is recognized that financial issues for both 
partners are more complex than ever before. I hope that we can 
jointly commit to ensuring that the research infrastructure is 
maintained, nurtured and permitted to evolve along with the 
science that it supports.
    I greatly appreciate the opportunity you have provided to 
me to present this information, and that concludes my summary.
    [The statement of Dr. Horvath follows:]

                Prepared Statement of Albert G. Horvath

    Chairman Lipinski, Ranking Member Ehlers, and other distinguished 
Members of the subcommittee on Research and Science Education, thank 
you for allowing me to participate in this hearing on a topic that is 
very important to those of us who manage the financial and 
administrative aspects of organized research at major research 
universities.
    My testimony is provided on behalf of the Pennsylvania State 
University, where I am Senior Vice President for Finance and Business/
Treasurer, and representing the Council on Governmental Relations, or 
COGR, where I currently serve in the role of Chairman of the Board of 
Directors. COGR is an association major research universities (and 
affiliated academic medical centers and research institutes) that helps 
to develop policies and practices that fairly reflect the mutual 
interests and separate obligations of federal agencies and universities 
in research and graduate education.

Background and Context

    As a chief financial officer of a major research institution, 
fiscal oversight of the research enterprise is an important and 
challenging aspect of my responsibilities. Research activities account 
for almost 19% of Penn State's operating budget, trailing only 
instruction and patient care in scale. The percentage of total revenues 
generated by research could be much more significant at other 
universities, depending on their mix of various mission-driven 
activities, Research is a complex activity which requires dedicated 
facilities, specialized equipment, significant physical infrastructure, 
substantial administrative support, and a number of specific compliance 
processes.
    While research is central to our universities' missions, keeping 
the research enterprise solvent, and keeping our finances solid, has 
become a greater challenge. Since the early 1990's, universities have 
faced tighter regulation with respect to funded research, with more 
limitations on our ability to effectively and reasonably recover those 
costs. A COGR paper in 2004 entitled ``A New Research Business Model-
Incentivizing Universities'' (Attachment B) stated:

     ``Six examples describe the impact of current regulations, all of 
which provide short-term cost savings to the federal government, at the 
risk of long-term damage to the research enterprise. The regulations 
impose:

          Limits on legitimate cost recovery by agency or type 
        of award,

          A cap on administrative cost recovery in a time of 
        growing administrative and regulatory requirements,

          Lack of commitment to life cycle costs for capital 
        projects and the requirement to invest capital recoveries,

          An artificial distinction between internal and 
        external interest costs on borrowed funds,

          The exclusions of many universities from receiving 
        adequate utility cost reimbursement,

          Conflicting and duplicative requirements among 
        funding agencies.''

    Universities have had to subsidize the physical and administrative 
infrastructures supporting research with revenues generally provided by 
State governments (public universities) or private philanthropy 
(private universities). Following the economic challenges of the past 
18 months, both of those funding sources have become seriously 
constrained.
    Additionally, the infrastructure necessary to enable cutting edge 
research is complex and expensive. Universities have made significant 
investments in such infrastructure--buildings, major equipment, utility 
systems, organizational changes and processes with long-term financial 
commitments based upon an expectation that research funding would 
continue over the term of amortization. Regulated funding restrictions 
and budget uncertainty conspire to create tremendous financial risk and 
anxiety over the ability to fund the debt that has been incurred. It 
also dampens the willingness to make new investments in the future.
    A summary of Penn State's research funding/activity can be found in 
Attachment A.

How research infrastructure is financed at Penn State

    The University's overall capital plan is financed through a variety 
of sources. The current multi-year plan, which runs through June 30, 
2013, anticipates a total of $820 million of projects that will be 
financed as follows:




    Major research facility projects--renovations or new construction--
are generally enabled through the issuance of tax-exempt bonds. This 
long term obligation (20-30 year repayment) is repaid using 
unrestricted revenue sources, including a portion of recovered facility 
and administrative costs (F&A) on sponsored awards. The Commonwealth of 
Pennsylvania, in addition to its annual $40 million commitment of 
capital funding, sometimes provides special allocations for such 
facilities, but these commitments historically amount to a modest 
fraction of total construction cost.
    Penn State has a strong credit rating (AA by Standard & Poors, Aa2 
by Moody's Investor Services) and has been successful at obtaining 
favorable interest rates for its tax exempt bond issuances. This access 
to ``reasonably priced'' funding has enabled the University, along with 
many other research institutions, to invest in its facilities and 
infrastructure particularly as endowments grew and balance sheets 
strengthened. Encouraged by the commitment to research funding, 
particularly through NIH during the 1990s, research facilities were 
expanded or renovated to enable the cutting edge work implied by such 
Federal investments. However, a few critical changes have caused 
uncertainty and anxiety around the sustainability of these expensive 
and complex assets.

        1.  Once the doubling of the NIH budget was achieved, 
        subsequent years' budgets began to erode some of that growth. 
        While there was no assumption that such extraordinary growth 
        would continue forever, allowing some of that growth to recede 
        was not expected. This has caused uncertainty in planning for 
        the future and put pressure on universities' ability to service 
        its external debt as had been planned.

        2.  The severe economic downturn, which hit higher education 
        beginning late in 2008, has significantly reduced endowment 
        values and constrained future borrowing potential. The ability 
        to continue to reinvest in research at past levels will be 
        difficult if not impossible, given that those facilities must 
        compete for priority against all other activities of 
        comprehensive universities (classrooms, student support 
        facilities, libraries, student housing and the like). While 
        markets and endowments have somewhat recovered, it will take 
        some time and sustained improvement for values to return to 
        what they were previously.

        3.  Other sources of possible funding for research facilities--
        private philanthropy and state investments--have fallen 
        considerably because of the effects of the economic downturn.

    The events noted above have dampened the ability of research 
universities to invest in all facilities, including those supporting 
research. The specific impact at Penn State has been a delay in our 
ability to move ahead with our five year capital plan as originally 
drafted. Projects generally have not been cancelled, but many have been 
delayed by generally 12-18 months. Also, our borrowing plans, although 
historically conservative, have become even more so as we monitor the 
activity of the capital markets and move cautiously with new debt.
    As an example of the process followed and the issues encountered 
with the planning and execution of a major research facility is Penn 
State's Millennium Sciences Complex. This 175 thousand square foot 
research facility will house faculty conducting research in material 
science and the life sciences. The building is intended to encourage 
collaboration between these two disciplines and will include many of 
Penn State's most pre-eminent research faculty, The building is 
projected to cost $215 million and is scheduled for completion in 2011. 
The Commonwealth has provided $82 million of funding (the majority of 
which was a Penn State allocation of its annual capital allocation); 
the $133 million balance is financed with bonds issued in 2009. The 
interest expense associated with this project is calculated to be $63.8 
million over the life of the bonds. Sponsored research funding 
generated by faculty in this facility will provide partial repayment of 
the interest costs of the related borrowing through recovery of F&A 
costs.
    Penn State has approximately 1.6 million square feet of space 
dedicated to research at its University Park campus, with expressed 
needs for up to another 500 thousand square feet just to support the 
present research portfolio. There are over $32 million of identified 
research equipment needs, Over $475 million of deferred maintenance 
exists in its research buildings, based upon facility condition audits 
conducted across the campus. Other research needs/initiatives could be 
addressed with additional facilities that don't currently exist. 
Clearly, this is a big challenge as these needs must compete for access 
to funding against other institution priorities (including the Penn 
State Hershey Medical Center, which also competes for this investment 
capital and supports a significant research enterprise).

Academic Research Infrastructure Program (ARI)

    The inclusion of additional funding for the ARI as part of the 
American Recovery and Reinvestment Act was most welcome in the research 
community, While the total amount allocated to funding for renovation 
and renewal of existing facilities, this was a positive step toward 
helping research universities to address a critical issue--deferred 
maintenance and aging facilities, Without a regular stream of funding 
toward such buildings and equipment, they become obsolete. Many labs 
exist in buildings built 30 or more years ago. Building systems did not 
contemplate the requirements of modem day science and engineering 
research. Often the most difficult part of recruiting new research 
faculty is the extent to which facilities need to be upgraded or 
renewed in order to support the research program of the faculty member. 
Such upgrade can run into the hundreds of thousands or millions of 
dollars.
    Both Penn State and COGR would encourage the Congress to consider 
extension of the ARI in future years to assist in dealing with the 
challenge of maintaining facility viability. It will help in generating 
positive economic activity as well.
    The trade-off between increases in direct research funding versus 
more money in the ARI or other infrastructure support programs is a 
difficult one. ARI has been most welcome and beneficial. However, a 
strong, consistent stream of funding for the primary research 
supporting agencies is also critical, This basic research funding 
provides the support for labs, research technicians, graduate students, 
support personnel, as well as funding toward supporting infrastructure. 
We hope that consideration of accomplishing both of these goals--steady 
strong research funding and some form of ARI--could be goals that are 
achievable together and not at each other's expense. Additionally, we 
would encourage establishment of a larger fund for such projects. As 
noted earlier in my testimony, the needs for such funding are large and 
compelling. The ability to fund a larger number and wider range of 
projects would be extremely effective in maintain facility capability.

Other ideas to provide support for research infrastructure needs

    We would encourage the consideration of additional investments by 
the Federal government to help support the infrastructure that supports 
research at universities. Such investments will help to ensure that the 
continued cultivation of the basic science as the fundamental 
foundation of innovation and progress envisioned by Vannevar Bush 
several decades ago. Also, the economic benefit that such research 
provides is demonstrable. However, recognizing the realities of 
difficult budget choices, I offer some other ideas for ways in which 
the Federal government can more effectively support research on 
campuses and reduce some of the growing burdens that this activity 
places on University finances.

          Predictable, long term research financing--By far, 
        the most important element toward reducing risk as universities 
        make substantial financial commitments to research, spanning 
        several years, would be a reliable stream of direct research 
        revenue. As budgets are prepared and business plans developed, 
        a major assumption in the evaluation of a project is the 
        reliability of revenues that will be available to repay debt 
        incurred. If a long term commitment to predictable growth in 
        major funding agencies budgets for extramural research, it 
        would help to provide confidence that the research such 
        facilities are designed to support will, in fact, be able to 
        financially support them over a 20-30 year period.

          Federal programs to assist in financing research 
        infrastructure--The Federal government could provide mechanisms 
        to help reduce cost of major investments or the risk of long 
        term decisions. Such programs/systems would help to incentivize 
        new investments in an era following balance sheet declines. The 
        economic benefit of such investments has both a short run 
        component (the activity stimulated during the construction 
        phase) and a long run component (addition of higher paying 
        knowledge jobs).

                  Provide support for research capital investments

                        `  Pool of capital dedicated to support 
                        investments in research buildings, building 
                        renovations, computing infrastructures

                                  Explore the possibility that 
                                a pool of dedicated capital could be 
                                made available to research universities 
                                with very favorable repayment terms. 
                                These funds would be accessed by 
                                institutions, according to specific 
                                criteria (types of facilities/uses) to 
                                finance new facilities and would be 
                                repaid at subsidized interest rates. 
                                This pool then would be self-sustaining 
                                over time and could help to ease some 
                                of the tough choices universities face 
                                on how to invest its limited capital.

                        `  Debt service subsidies for university-issued 
                        bonds

                                  Provide subsidies for payment 
                                of debt service on borrowing undertaken 
                                by research universities for new 
                                facilities, major renovations or major 
                                equipment purchases that benefit 
                                Federal sponsors.

                  Provide ``grants'' to fund a portion of new 
                facilities, major building renovations or capital 
                equipment acquisitions, This would serve to reduce the 
                overall amount needed from bond issues or other 
                external borrowings, thus reducing the impact on an 
                institution's credit rating/debt capacity issues.

          Allow recovery for the cost of internal capital, 
        which is permitted for commercial contractors

                  There are a number of differences between the cost 
                accounting rules that exist for commercial and non-
                profit recipients of Federal funds. One notable example 
                is the inability for universities to recover the 
                ``cost'' of internal funds (reserves) that are used to 
                finance research assets. Changing OMB Circular A21 to 
                allow such costs to be recovered would help to incent 
                perhaps a greater commitment of institutional reserves 
                into such projects. This cost would become a component 
                of the institution's F&A rate, which is audited and 
                approved by its cognizant Federal agency.

          Consider elimination of the administrative cost cap 
        in OMB Circular A-21

                  While not directly related to ``bricks and mortar,'' 
                the cap that was placed upon administrative cost 
                recovery by universities almost 20 years ago continues 
                to create burdens on institutions. Most of the major 
                research institutions have calculated administrative 
                cost components which exceed the 26% cap, resulting in 
                millions of dollars worth of legitimate research 
                support expenses that go unreimbursed. No other type of 
                contractor performing work for the Federal government 
                is subject to such a cap on supportable, allocable 
                support costs. Since implementation of the cap, several 
                new requirements and regulations have been enacted that 
                require greater effort by universities; however, none 
                of the incremental costs associated with the regulatory 
                changes are recoverable (if the institution is over the 
                administrative rate cap).

                  If removal of the administrative cost cap is not 
                considered feasible, then consideration of the creation 
                of a new, uncapped pool for regulatory compliance costs 
                should be considered. As mentioned above, the growth of 
                new requirements, seemingly every few weeks, has placed 
                financial pressure on universities which other non-
                research revenues must subsidize. Universities WANT to 
                be compliant,, and often the regulations are complex, 
                requiring new investments or additional staff. This is 
                not a signal of inefficiency but recognition of the 
                cost of being compliant. A listing of new compliance 
                requirements implemented since imposition of the 
                administrative cost cap was compiled by COGR in March 
                2009, and demonstrates this point effectively 
                (Attachment C).

          Require all Federal agencies to reimburse 
        universities at Government-approved F&A, rate--A study by the 
        Rand Corporation in 2000 estimated that universities are 
        subsidizing Federally sponsored research by roughly $0.7 
        billion and $1.5 billion that would be eligible for 
        reimbursement through negotiated/approved F&A rates if all 
        sponsors were paying the approved rate. We would support 
        changes in appropriate regulation that would require Federal 
        sponsors to pay the negotiated Federal rate on all research it 
        funds.

Conclusion

    The research partnership between the Federal government and U.S. 
research universities has enabled great achievements in science, 
innovations that have fueled economic growth, and helped our system of 
higher education to become the envy of the world. It is recognized that 
financial issues for both partners are more complex than ever before. 
Infrastructure--in the form of buildings, equipment, computing 
networks, and other necessary support ingredients--allow this research 
to flourish and discoveries to be made. We must jointly commit to 
ensure that this infrastructure is maintained, nurtured and permitted 
to evolve along with the research that it supports. As discussed in the 
2003 COGR paper ``New Research Paradigms Call for Regulatory Change:'' 
(Attachment D).

     ``The essential premise for a new business relationship between 
the government and universities is the simple acknowledgment that both 
parties engage as ``business partners''. This means, among other 
things, a recognition of complementary interests in the cost effective 
administration of awards and the providing of adequate funds to meet 
the joint expectations for the outcomes of research.''

    Steady, predictable streams of research funding form the foundation 
for the science and technology discoveries that result. A commitment by 
the Federal government to such funding will help to make future 
investments in facilities and support by universities less risky 
projects. Additionally, the continuation of programs like the Academic 
Research Infrastructure Program, NSF's Major Research Instrumentation 
program, among others will assist in the development of new facilities 
along with ensuring the viability of existing facilities. Finally, 
changes to policies that will enable full and reasonable recovery of 
costs associated with research will help to ease pressure that caps and 
other funding reductions have created.
    I greatly appreciate the opportunity you have provided to present 
this information.

Attachment A









Attachment B

A New Research Business Model: Incentivizing Universities

Introduction

    This paper illustrates some of the difficult choices that 
universities face in order to comply with the current cost policy 
restrictions of the federal government as expressed by federal agencies 
and by OMB. Many of these policies were imposed for the purpose of 
saving taxpayer dollars, while at the same time providing maximum 
support for research. The paper illustrates that these measures are not 
the best way to reach the stated goals. Instead of imposing 
restrictions, we propose that the government look to a new business 
model, We maintain that it would be preferable to offer incentives to 
allow the academic community to manage the increasingly rare research 
finds on terms comparable to those granted to any experienced, cost-
conscious and competitive provider of research.
    Six examples describe the impact of current regulations, all of 
which provide short-term cost savings to the federal government, at the 
risk of long-term damage to the research enterprise. The regulations 
impose:

         Limits on legitimate cost recovery by agency or type of award,

         A cap on administrative cost recovery in a time of growing 
        administrative requirements, Lack of commitment to life cycle 
        costs for capital projects and the requirement to invest 
        capital recoveries,

         An artificial distinction between internal and external 
        interest costs on borrowed funds, The exclusions of many 
        universities from receiving adequate utility cost 
        reimbursement, Conflicting and duplicative requirements among 
        funding agencies.

    Taken either in isolation or as a short-term mandate, any one of 
these restrictions may seem immaterial, but their cumulative impact is 
acutely felt in the higher education community. The costs to the 
federal government are not as immediately evident but nonetheless real. 
They may be financial, they sometimes represent lost opportunities, 
i.e. loss of research capacity or they inadvertently encourage less 
than optimal research decisions. The impact would be no different and 
as strongly resisted, if these regulations were imposed on any other 
business entity.
    It is true that the government's current costing polices do not 
deprive universities of the freedom to make choices, but the choices 
have become increasingly narrow and at times dysfunctional. It should 
be evident that universities can best serve the nation if they are free 
to make their best intellectual judgment about the future direction of 
research. Universities have historically accepted the risk that today's 
research focus is likely to change tomorrow, if new discoveries 
redirect the progress of science or if the federal government changes 
its priorities. The risk, however, is much greater today, because 
artificial limits and prescriptions have been injected into the basic 
compliance structure that is the foundation of government research 
support policy. Cumulatively these prescriptions create financial 
disincentives that impact program choices and may encourage decisions 
that are neither in the best interest of taxpayers' nor consistent with 
government goals and policies in supporting university research.

Situation 1: Limits on legitimate cast recovery by agency or type of 
        award

    A tenured professor retires and her 30-year-old laboratory must be 
renovated to accommodate a new faculty member. In deciding how to 
allocate the new space, the university has to weigh the risk that a 
junior faculty's primacy research support may come in the form of 
awards that do not pay the full, negotiated facilities and 
administrative (F&A) rate. An example is the NIH career development 
program, under which F&A recovery is limited to 8%. In establishing 
this program the government has demonstrated its resolve to support 
young investigators, recognizing that the nation needs a steady stream 
of accomplished researchers. However, the government's failure to 
provide the requisite support costs diminishes the impact of the 
program. Alternately, a more senior faculty member, whose cutting-edge 
research is so novel that federal funding has yet to be secured, might 
claim the newly renovated space. But for how long can the university 
afford to cost share the F&A costs for his laboratory?
    The financial realities of setting up a recent hire to the faculty 
are large and are real. Renovating space, equipping and running a modem 
research lab, and providing seed money for new research requires as 
many institutional resources (research proposal assistance, utilities, 
administrative support) as providing research space for a more senior 
colleague. It may turn out that the space is given to an established 
investigator operating in a mature branch of her field, where F&A costs 
are provided. Financial considerations, of course, may not be the final 
determining factor in. academic decisions. Nevertheless, from a 
strictly ``business'' perspective, the choice favors the more senior 
appointment in ``safe'' and well-funded research areas.
    These situations illustrate how academic and financial arguments 
create difficult trade-offs when important strategic research decisions 
are made. Some universities have decided to recognize faculty, who are 
successful grantees and who bring in the full negotiated F&A rate, by 
awarding them the best research space in newly-renovated buildings. 
Other universities might consider prioritizing the renovation of 
buildings that will house successful grant applicants, at the expense 
of other disciplines that are not as likely to be ``funding winners.'' 
Without sufficient revenues to support total costs, both direct and 
indirect, the need quickly arises to subsidize research activity from 
other sources (primarily from tuition or gifts from private donors).
    These cases illustrate how financial disincentives may lead to 
results that are neither in the best interest of the taxpayers nor 
consistent with government goals and policies in support of university 
research.

Situation 2: A cap on administrative cost recovery in a time of growing 
        compliance requirements

    The university determines that changes to its oversight 
requirements for bio-medical research must be substantially enhanced 
due to much stricter regulation by the federal government. Such 
enhancements will require the institution to incur additional costs in 
the hundreds of thousands or even millions of dollars, all directly 
supportive of research primarily sponsored by federal agencies. Some 
are the result of new regulations, others are the result of changes in 
federal interpretation of compliance needs in research areas that 
develop commensurate with the advance of scientific methods. However, 
due to the fact that the institution's negotiated F&A rate is already 
at the 26% administrative cost cap, the government does not provide its 
fair share of these new compliance costs.
    In order to continue the relevant research and to comply with 
federal regulations, the administrative costs supporting such research 
must be subsidized by other revenue sources, possibly tuition revenue 
or private donations or increasingly shrinking state funds. Compliance 
with these federal mandates is not a matter of choice and 
``efficiencies'' cannot be easily achieved. If no revenue sources are 
left and if additional subsidies to the research enterprise can no 
longer be found, the research-intensive universities face even more 
difficult choices. Their options are reduced to gradual elimination of 
certain research programs or the even more problematic decision of 
reducing compliance standards to the required minimum level.

Situation 3: Decreasing pool of investment capital for universities to 
        invest in research facilities

    The university determines that a research building dating from the 
I930s has outlived its usefulness and that it would be most cost-
effective to raze the building and to build all new space. However, the 
common source of funding in the past-tax-exempt bond funding-is now in 
shorter supply. With the economic downturn of the past few years, the 
institution's endowment value has declined, thus reducing the level of 
borrowing capacity available. While the institution has extensive 
borrowing ability, a significant additional borrowing will likely push 
its credit rating down a level, thus creating more expensive interest 
costs. Capital campaigns, although desirable, can only be initiated 
slowly and have unpredictable outcomes. These realities could force the 
university to defer a decision, with the result that the building in 
question will deteriorate further and that current or future research 
will be negatively impacted.
    The federal government has for years failed to include a well-
funded facilities support program in the federal agency research 
program budget. The unfortunate consequence of this has been the 
increase of earmarks in Congressional appropriations. Also, to date 
there exists, no federal policy that commits the government to 
participate in its share of debt service over the life of the loan. It 
would make good business sense to provide incentives, since the 
research to be performed in these facilities is in the government's 
interest. A variety of incentives have been proposed and could include 
a reasonable facilities funding program placed in the federal budget, 
offering universities a federal sharing in loans, or a federal loan 
guarantee program. The common denominator in all these would be 
removing the uncertainty that universities currently experience under 
the given federal policy.
    Business decisions for the university are further complicated by 
the federal government's requirement that for every dollar of 
depreciation recovered on the new building, a dollar must be spent on 
some future project. This means in effect that the university will 
never recover the cost of its investment, and may be committed to new 
construction at times when it is not in a sound financial position to 
do so.

Situation 4: An artificial distinction between internal and external 
        interest costs on borrowed funds

    The university's bond rating is in jeopardy due to depressed 
financial markets. Because of this, it would be less expensive on a 
gross basis to allocate internal capital funds to pay for a new science 
building. However, the university knows that there is no option for 
recovery from the government of the university's internal cost of 
capital. Under these circumstances, the university may make a rational 
business decision to borrow at a higher external rate because it then 
can recover a fraction of the interest costs from the federal 
government, thereby lowering its net cost of interest.
    It would clearly be in the taxpayers' interest to provide 
incentives for the university to use its own money, possibly by sharing 
some of the investment costs, so that the federal contribution does not 
go to defraying avoidable higher interest costs.

Situation 5: The exclusion of certain universities, from receiving the 
        utility cost adjustment factor

    A space vacancy occurs in the cancer research center. The 
university could make the space available for a number of equally 
worthy projects. One of them is a large laboratory where the research 
project requires constant air changes. Unfortunately such a laboratory 
would entail high energy use. Since the university had not undertaken 
an energy study prior to 1996, it is now prevented from receiving 
higher compensation through the F&A rate for higher energy research 
consumption. When the government put new energy studies on hold in 
1996, it promised to develop a fair formula for all academic energy 
consumers, but has yet failed to do so. As a result, more than one 
hundred universities are now prevented from recovering the costs for 
higher energy use which they consume, which they could easily document, 
and which their peers who had done prior energy studies now enjoy.
    For the university faced with this choice, the uncompensated energy 
costs inject an artificial economic factor into a determination which 
should be based solely on academic and scientific needs. This may 
influence the university to make a decision not in the best interests 
of science.

Situation 6: Regulatory Burden Reinforces Bureaucracy

    Universities, like commercial businesses, seek to maximize the use 
of financial and human resources to meet their strategic objectives. As 
resources become more constrained, institutions attempt to streamline 
their efforts, including the maximum use of available technology, to 
reallocate resources into priority areas. But research universities 
find it increasingly difficult to reduce the administrative burden of 
research in order to fund more strategic activities. The following 
three examples illustrate how federal requirements contribute to high 
administrative costs, and thereby detract from effective decision-
making.

          Each federal agency views itself as unique, operating 
        within its own set of administrative guidelines and 
        regulations. This requires counterpart experts at the 
        university to effectively deal with the day-to-day operations 
        of each respective agency funding.

          New initiatives in the government's administrative 
        processes, such as electronic research administration, have not 
        been coordinated across sponsoring agencies in a way that 
        establishes common standards to assist universities in 
        implementing simple, cost effective solutions. As a result, the 
        same task (i.e., submission of a research proposal) must be 
        accomplished in a variety of different ways.

          Costing regulations became more burdensome and 
        complex in the 1990s, including the requirement to develop and 
        adhere to Cost Accounting Standards Board protocols that are 
        expensive to implement and maintain but often ignored by the 
        cognizant agencies. Significant time and effort that could have 
        been used elsewhere was expended to develop the Disclosure 
        Statements, and continues to be devoted to maintaining them as 
        changes are required to business processes.

    A specific example of how such administrative requirements 
negatively impact the drive towards efficiency is one university's 
attempt to create a new staff classification for ``research faculty. `` 
The sole motivation for this change is to build and strengthen the 
university's research capability. In order to create the new position, 
the institution needs to have a cost recovery model that enables all 
leave costs to be charged directly to grants and contracts. This 
requires changes to the written disclosure statement that must be 
negotiated through the cognizant agency, even though costing practices 
were recently reviewed in negotiation of the F&A rate. Such 
requirements delay the implementation process in a way that is 
difficult to explain to faculty administrators and counter to the 
shared desire for increased efficiency.

Conclusion:

    These examples illustrate the dangers to the research enterprise 
when sound business decisions can no longer be reconciled with what 
appear to be sound research decisions. Unless the government changes 
some of its policies, it may trigger outcomes that are adverse to its 
own stated goals: to foster an environment in which the government's 
academic business partner is empowered to manage itself and the federal 
investment in research in the most pendent and cost effective manner. 
For each of the examples cited in this paper, there exist. a number of 
possible policy solutions. Some are very specific, such as extending 
the utility cost adjustment factor, and allowing ``cost of money'' for 
universities that use their own capital for facilities. Others have 
broader implications, such as eliminating the administrative cap, or 
establishing loan guarantee programs for new facilities. Still others 
would only require adherence to government-stated purposes and 
principles, such as payment of negotiated F&A rates on all federal 
awards, and streamlining financial and research administration 
requirements to eliminate unnecessary processes and to create 
consistency among sponsoring agencies.
    The unique position of universities precludes their being treated 
as business partners in the same sense as commercial business partners. 
According to the latest information from NSF, university funds to 
support research' reached $6.55 billion in 2001, or 20% of total 
expenditures, As recently as 15 years ago the university share of total 
expenditures was 10%. Stated differently, for every $1 million of 
funding received in new awards, a university provides an additional 
$200,000 of direct and infrastructure costs. This significant 
investment of funds demonstrates the universities' commitment to 
support research. Recognizing this, universities must and do take every 
opportunity to maximize administrative efficiency and reduce costs. 
However, we believe a comprehensive strategy is needed to address the 
growing imbalance in support for the university research 
infrastructure. The government will justifiably expect to get the best 
results from federally funded research on the most reasonable terms, 
and with the expectation of cost sharing. But where the government 
fails to recognize the universities' legitimate business constraints, 
and the result is increased research cost shifting, it is time for new 
research business models, which recognize these constraints and design 
funding programs consistent with them.

Attachment C

Federal Regulatory Changes, Since 1991
These regulations directly affect the conduct and management of 
research under Federal grants and contracts. The list of current 
regulations is in chronological order.

Federal Policy for the Protection of Human Subjects (Common Rule, 1991)

Nonindigenous Aquatic Nuisance Prevention & Control Act of 1990 
(Implemented, 1992)

NIH Guidelines for Research Involving Recombinant DNA Molecules (1994)

Deemed Exports (1994, EAR & ITAR)

         DFARS Interim Export Control Compliance Clauses (July 2008)

Conflict of Interest

         Public Health Service/NIH Objectivity in Research (1995)

         NSF Financial Disclosure Policy (1995)

Lobbying Disclosure Act of 1995

Cost Accounting Standards (CAS) in OMB Circular A-21 (1995)

Health Insurance Portability & Accountability Act of 1996 (HIPAA) 
Privacy Rule

OMB Elimination of Utility Cost Adjustment (UCA) (1998)

Data Access/Shelby Amendment (FY 1999 Omnibus Appropriations Act); 
related amendments to OMB Circular A-110

Policy on Sharing of Biomedical Research Resources (NIH, 1999)

Misconduct in Science (Federalwide Policy, 2000)

         NEH, 2001

         NSF, 2002

         EPA, (Directive, 2003)

         Labor, 2004

         HHS/PHS, 2005

         NASA, 2005

         Energy, 2005

         Veterans Affairs, 2005

         Education, 2005

         Transportation, 2005

         USDA (Proposed, 2008)

HHS Centers for Medicare and Medicaid Services (CMS) National Coverage 
Determination for Routine Clinical Trials (Clinical Trials Policy), 
2000

Executive Order 13224, Blocking Property and Prohibiting Transactions 
With Persons Who Commit, Threaten to Commit or Support Terrorism 
(September 2001, also EO 12947, 1995)

Select Agents & Toxins (under CDC and USDA/APHIS) Public Health 
Security & Bioterrorism Preparedness & Response Act of 2002; companion 
to the USA PATRIOT Act (2001)

FISMA Federal Information Security Management Act (Title III, E 
Government Act of 2002) OMB Circular A-130, Management of Federal 
Information Resources, Appendix III, Security of Federal Automated 
Information Systems

CIPSEA Confidential Information Protection and Statistical Efficiency 
Act (OMB Implementation Guidance 2007, Title V, E Government Act of 
2002)

Federal Policy on Embryonic Stem Cell Research (2003)

Data Sharing Policy (NIH, 2003)

Homeland Security Presidential Directive (HSPD)-12, Common 
Identification Standards for Federal Employees and Contractors (2004)

Higher Education Act, Section 117 Reporting of Foreign Gifts, Contracts 
and Relationships (20 USC 1011f, 2004)

Model Organism Sharing Policy (NIH, 2004)

Constitution & Citizenship Day (2005, Consolidated Appropriations Act 
FY 2005)

Genomic Inventions Best Practices (2005)

Combating Trafficking in Persons (2008)

Code of Business Ethics & Conduct (FAR) 2008

Homeland Security Chemical Facilities Anti-Terrorism Standards (CFATS) 
2008

E-Verify 2008

Military Recruiting and ROTC Program Access (2008, Solomon Amendment, 
National Defense Authorization Act for FY 2005)

Nuclear Regulatory Commission Order Imposing Fingerprinting and 
Criminal History Records Check Requirements for Unescorted Access to 
Certain Radioactive Materials (Feb. 2008, Section 652, Energy Policy 
Act of 2005)

National Institutes of Health Public Access Policy (2008, Consolidated 
Appropriations Act of 2008, Division G, Title II Section 218)

Certification of Filing and Payment of Federal Taxes (Labor, HHS, 
Education and Related Agencies Appropriations Act of 2008, Division G, 
Title V, Section 523)

Health and Human Services/FDA Clinical Trials Registry Implementation/
Interpretation Changes, Since 1991

Foreign Nationals (See COGR/AAU/FDP Troublesome Clause Report, 2008 
\1\)
---------------------------------------------------------------------------
    \1\ The Report is available at: www.cogr.edu/docs/
COGRAAUTroublesomeClausesReport.pdf

---------------------------------------------------------------------------
Publication Restrictions (see COGR/AAU/FDP Troublesome Clauses, 2008)

P.L. 106-107/Grants.gov: Electronic Applications, Financial Reporting 
Progress Reports, iEdison Invention Reporting, etc. CCR/DUNS Registry 
requirements

Subrecipient Monitoring (OMB Circular A-133, Compliance Supplement)

Changes to A-21 F&A Proposal Format

Federal Policy for the Protection of Human Subjects:

         Federalwide Assurance (2004), mandatory training

         IRB Registration (2008)

Title IX of Education Amendments of 1972: Access to science and math 
educational programs (2007+)

EPA Hazardous Waste, Subpart K (2008)

IRS 990 Reporting

Significant Proposed Changes

Export Controls: Export Administration Regulations (EAR) & 
International Traffic in Arms Regulations (ITAR) (2003)

Responsible Conduct of Research Training--NSF (America COMPETES Act 
2006)

Federal Funding Accountability and Transparency Act (FFATA) 
Subrecipient Reporting (2006)

National Science Advisory Board for Biosafety (NSABB) Oversight of Dual 
Use Life Sciences Research of Concern

Nuclear Regulatory Commission--Considerations concerning the Security 
and Continued Use of Cesium-137 Chloride Sources (July 2008)

USAID Partners Vetting System (re: EO 13224 et al. re: terrorist 
financing)

USDA Animal Welfare Act, Contingency Planning (2008)

Attachment D

           New Research Paradigms Call for Regulatory Change

Executive Summary

    During recent discussions initiated by the Office of Science and 
Technology Policy about new research business models, much attention 
was given to interdisciplinary research activities and the team efforts 
required to carry out such research. Expanding these thoughts further, 
this paper offers an analysis of the increased administrative 
responsibilities that are encountered when research projects scale up 
to more complex, multi-disciplinary and multi-institutional levels.
    Starting with streamlining that would benefit the administration of 
the basic assistance award, the paper recommends changes that would 
facilitate business practices commensurate with increasingly complex 
business relationships. The requested changes in current federal 
regulations described here are not new, but are gaining greater urgency 
in order to assure accountability and to reduce the administrative 
burdens and costs that impact both the government and its awardees as 
projects scale up.
    Although this paper focuses primarily on the government-university 
relationship, it does not seek to diminish the importance of 
university-industry collaboration nor does it deny the many beneficial 
relationships existing between universities and their State agencies. 
The close nexus between education and research that exists in 
universities makes them not more important, but certainly different 
from most other research providers.

Introduction: Premise for research business relationships

    The essential premise for a new business relationship between the 
government and universities is the simple acknowledgment that both 
parties engage as ``business partners''. This means, among other 
things, a recognition of complementary interests in the cost effective 
administration of awards and the providing of adequate funds to meet 
the joint expectations for the outcomes of research. These mutual 
interests exist in both the assistance and the procurement mode because 
in each both parties provide value. Towards these ends, regulatory 
requirements that create unnecessary burdens should be removed, and 
funding for administrative expenditures should be based on a thorough 
and fair examination of the universities' F&A documentation. The term 
``rate negotiation'' is inappropriate and implies a broken process. 
Equally important is the avoidance of cost shifting and imposing of 
caps and other restrictions by the government, which the commercial 
sector would describe as ``price controls''. It now appears that not 
only has the Congress called for new business practices, as evidenced 
in P.L. 106-107, but that the White House, through the Office of 
Science and Technology Policy has joined that call for change.

Business Models for the Basic Assistance Award

    The simplest research platform is a basic assistance award, which 
may provide research support of up to $1 million in federal funding. An 
example of how a new business practice could remove unnecessary 
regulatory burden for even this simple platform is provided by the 
proposals of Robert Newton, a former NSF official.
    In. the early 1980s, Newton proposed that a faculty's entire 
research should be considered as one ``research program'' to be managed 
as an integrated whole rather than as individually sponsored and 
managed ``research projects''. The key prerequisite to aggregation was 
the concept of ``relatedness'', which the faculty researcher would be 
obligated to assert and demonstrate. Once relatedness was established, 
the researcher should be able to use all sources of funding to charge 
costs to serve research needs rather than be restricted by individual 
agency budgets. This concept was one of the motivating factors for 
forming the Florida Demonstration Project in 1983. It is not yet widely 
embraced in the Federal Demonstration Partnership of 2003.
    Several other unnecessary regulatory impediments to the cost-
effective management of research could be similarly eliminated by 
simple changes to the current requirements. These include flexibility 
in starting a project, the ability to adjust expenditures according to 
the needs of the research without having to obtain agency prior 
approval for each individual action; and the authority to extend the 
timeframe for expenditures as dictated by progress on the project, 
without being accused of violating the ``expenditure rate''.
    The value of the business efficiency of such changes was recognized 
when OMB revised Circular A-110 in the early 1990s. OMB directed 
federal agencies to adopt a unified position on grant management 
matters and to provide ``expanded authorities'' to the grant recipient 
for management without the need for individual prior agency approval at 
each step. This recommendation reflected broad public support. Agency 
implementation however was uneven and even today federal agencies are 
far from uniform in granting such ``expanded authorities'' to research 
universities. In 2000, public comment on the statutory requirements of 
P.L. 106-107 again indicated overwhelming support for granting expanded 
management authorities to all funding recipients under all government 
awards. Until these simple steps are taken to adopt sensible business 
practices, there is little point in discussing the more complicated 
issues associated with more complex research and funding efforts.

Recommendation:

          Revise OMB Circular A-110 to direct all federal 
        agencies to grant the ``expanded authorities'' for grants 
        management in accordance with federal regulations to all 
        research universities.

          Endorse the concept that individual but related 
        research projects by a single investigator can be considered 
        one research program for purposes of management and 
        accountability.

          Rely on business system reviews and project audits at 
        universities rather than prior approvals by individual federal 
        agency staff.

Business Models for Multi-Sponsor, Multi-University Projects

    Awards for multi-sponsor, multi-university projects range from $1-5 
million dollars. Coordination and leveraging of effort is critically 
important to their success. It is widely acknowledged that their 
management is complex because they involve teams of scientists working 
at different sites and on various aspects of one common research 
problem. Yet, in most cases, none of the participating universities has 
enough support to cover more than their minimum share of the 
administrative burden. Because the federal agencies take a narrow view 
of the budget categories under OMB Circular A-21, sufficient funds are 
not provided to support the secretarial and clerical personnel required 
for such. a sophisticated effort.
    Recently, some federal officials seemed to imply that a ``new 
business model'' would require that OMB Circular A-21 be withdrawn and 
fundamentally revised. We do not believe that such drastic cure is 
required. All that is required to meet research needs is to go back to 
Circular A-21 in its original form. That would delete a number of 
requirements which do not add value.
    Several other modest changes to OMB Circular A-21 would further 
advance businesslike management of research. The language of the cost 
circular needs to be coordinated with the management circular to avoid 
discrepancies. Universities should be granted use of the cost of money, 
which other business sectors currently use. University responsibility 
for monitoring their sub recipient awards must be limited to reasonable 
procedures focused on scientific program objectives. The government 
should not expect universities to serve as auditors on one another's 
projects. This becomes particularly important in multi-campus 
arrangements, where the designation of subrecipient vs. research 
partner may not be sufficiently well defined.
    Another federal agency practice contrary to sound business 
principles is that not all federal agencies pay the negotiated F&A 
rate. They cite various reasons, some programmatic, some historical. 
This uneven approach to what is intended to be the government-wide rate 
becomes particularly visible and detrimental in multi-agency awards. 
The resistance by several agencies to fully fund the negotiated. F&A 
rates of universities results in extensive under recovery of costs that 
the science projects can ill afford. Respective data have been provided 
by the Rand Corporation and more recently by COGR. It would be good 
business. practice for all agencies to scrub their policies, some of 
which date from the late 50s, and to eliminate restrictions to full 
rate reimbursement that have been carried forward without appropriate 
statutory justification.
    Large multi-campus research projects may require institutional cost 
sharing. The capacity of the participant universities to come up with 
such funds is dependent on many factors. One might surmise that federal 
oversight over cost sharing as well as general project administration 
on multi campus awards would be facilitated by Cost Accounting 
Standards. However, internal consistency rather than commonality is the 
major objective of these standards. CAS standards add no value to 
multi-campus or to single investigator awards. They are duplicative and 
unnecessary because they reiterate A-21 standards. Doing away with 
these clearly unnecessary requirements, which the federal government 
admitted it cannot meet in a timely manner, would result in cost 
savings both for the government and for the universities that would 
clearly benefit research. Eliminating CAS standards is overdue.
    We recognize that the government has legitimate interests in the 
establishment of ethics safeguards and multi-disciplinary and multi-
campus projects may provide special concerns in this area. A new 
business model for this platform would benefit particularly from agency 
implementation of the government-wide misconduct in science policy 
promulgated in 2000. We also ask that all federal agencies follow the 
lead of NIH and NSF and develop conflict of interest regulations.

Recommendation:

          Return to the original language of 0MB Circular A-21

          Allow the direct charging of secretarial and clerical 
        staff

          Provide full funding of negotiated university F&A 
        rates

          Reduce subcontract monitoring to reasonable levels

          Issue Government-wide ethics rules

          Rescind the CAS coverage for universities

Business Models for Large Center Awards

    Institutions which compete for large awards for Centers or for 
specialized institutes for up to $15 million must commit substantial 
infrastructure support, Such support depends largely on available cost 
sharing resources. The size and complexity of these awards creates a 
big gap in management and operation between these awards and the single 
assistance awards. Yet, the same policies govern both. No reasonable 
business practice would expect to run a multi-million dollar automobile 
company like a neighborhood small business enterprise. The current 
restrictions in OMB Circular A-21 make no such distinction and as a 
result many of the large universities are stretched to the limit of 
their fund raising capacity for improvements of the infrastructure and 
for planning needed new facilities.
    For such large projects, the recovery of F&A costs is especially 
significant and consequently agency cognizance becomes a factor. 
Universities report considerable differences between the two cognizant 
agencies in their procedures for rate negotiation. There is no basis 
for two federal agencies to treat universities differently. Good 
business practices would call for closer coordination between DCAA and 
DCA, with respect to their audit and their oversight over F&A rate 
negotiation.
    The more one tries to scale up to a new platform, the clearer the 
impact of the cap on administrative cost will be felt. No other 
research performer is subject to caps, which were imposed in addition 
to three major revisions of the circular that took place in the 90s. 
While these OMB revisions provided a clearer definition of cost 
categories and eliminated ``gray'' areas, they also hold universities' 
administrative rates at a level that was below average even at the time 
it was adopted.
    After a decade without adjustment for cost increases, the 
university community is no longer able to cover the growing gap between 
regulatory requirements and the restriction in reimbursement. No other 
business is precluded by the government from recovering its legitimate 
business-related compliance costs. Since 1992, universities have had to 
absorb all administrative costs for new requirements and/or for the 
upgrading of systems that have become necessary in the intervening 
time. One would expect that it is in the government's own best interest 
to support universities in their effort to stay competitive and 
compliant, especially as new security measures become imperative for 
the nation. The cap needs to be lifted.
    These large awards also reinforce the need for government-wide 
acceptance of regulations governing human subjects, and to overcome the 
apparent distrust of the ``common rule'' which leads agencies to 
establish duplicative reviews of protocols and IRB procedures.
    Finally, unnecessary administrative costs could be eliminated 
simply by the establishment of government-wide payment systems that 
would replace the labor intensive and outmoded system of grant-specific 
draw-down by each federal agency.

Recommendations:

          Seek agreement between cognizant agencies

          Implement rate determination, not negotiation

          Remove the administrative cap

          Adopt a uniform government-wide payment systems

          Discourage duplicative administrative reviews

The New Research Business Model in Review

    As we propose it, the new business relationship between the 
government and universities is based primarily on trust. This trust 
relies on the understanding that it is in the university's own best 
interest to self regulate and hold costs down but also on the 
understanding that the government will provide stable funding and that 
the recovery of costs for facilities and for administrative services 
will not unexpectedly be capped.
    Universities face a growing perception that science can be 
separated from administration. That is a fallacy. Universities also 
witness the encroachment of administrative procedures that siphon funds 
that could otherwise support research or teaching. A new business model 
would eliminate such duplicative federal requirements.

Recommendation:

          Treat universities as business partners

          Permit performance based budgeting

          Set reasonable audit expectations

          Replace certifications with assurances

          Do not permit budget decisions to drive policy

    In this new business environment, the government will not be asked 
to appropriate more, it will merely be asked to allow universities to 
use resources the way universities determine necessary to support the 
mutual goal of obtaining the deliverable of sound scientific research.

In Conclusion

    Scaling up to different research platforms entails responsibility 
for scientific, administrative and financial decisions. It influences 
decisions regarding the workforce and infrastructure, including space 
and equipment, and calls for careful coordination between centers at 
different locations and subject to a variety of administrative 
regulatory requirements. It reaches into areas of regional and national 
security and raises fundamental questions regarding how one deals with 
potential restrictions on foreign scientists, with audit oversight and 
with the freedom to publish research results.
    The key to a successful research business model for increasingly 
complex projects lies in designing comprehensive but simplified 
administrative guidance and then permitting universities to take 
responsibility for management and oversight of the wide range of their 
projects. We believe that success depends largely on the extent to 
which the government will grant research universities the flexibility 
to make sound business decisions on campus, subject to subsequent 
review and audit by the government.

                    Biography for Albert G. Horvath

    Albert G. Horvath is senior vice president for finance and 
business/treasurer at Penn State effective July 1, 2009. He is 
responsible for leading the day-to-day management of Finance and 
Business and the strategic planning process for the unit which has an 
operating budget of more than $500 million and more than 2,500 
employees. He is also responsible for special projects and assignments, 
including information systems and technology, succession planning, and 
emergency preparedness.
    Al oversees the direct reporting relationships in the areas of 
auxiliary and business services; corporate controller and controller 
for the College of Medicine/Milton S. Hershey Medical Center; Office of 
the Physical Plant; University Budget Office; Office of Investment 
Management; Commonwealth Operations; University Police; and Human 
Resources.
    Al joined Penn State on June 29, 2007 as vice president for finance 
and business. He came to Penn State with a wide range of experience in 
finance and business, much of it in higher education. He has previously 
served as executive vice president for finance and CFO at Columbia 
University, where he has been responsible for the financial activities 
of the university--including its medical center--with a $2.7 billion 
operating budget. At Columbia, he developed a five-year capital plan 
and debt strategy and created a procurement organization, was involved 
with several issues at the medical center, and acted as primary 
administrative liaison to the audit and finance committees of 
Columbia's board of trustees. He also served as associate vice 
president for finance/controller, and later vice president for business 
and finance and CFO, at The California Institute of Technology; as 
controller at New York University; and as audit director and assistant 
vice president for finance at Carnegie Mellon University. He started 
his career as an auditor with Mellon Bank, before becoming a manager in 
Mellon's trust and investment department.
    A 1981 Penn State graduate with a degree in accounting, I also 
earned an MBA from Duquesne University in 1985.

    Chairman Lipinski. Thank you, Mr. Horvath.
    Dr. Raymond.

 STATEMENT OF DR. JOHN R. RAYMOND, VICE PRESIDENT FOR ACADEMIC 
AFFAIRS AND PROVOST, MEDICAL UNIVERSITY OF SOUTH CAROLINA, AND 
        CHAIR, STATE OF SOUTH CAROLINA EPSCOR COMMITTEE

    Dr. Raymond. Mr. Chairman and members of the Subcommittee, 
thank you very much for allowing me to testify today. The NSF 
EPSCoR program has a statutory function to strengthen research 
and education in science and engineering throughout the United 
States, and to avoid undue concentration of such research and 
education. This has been accomplished through providing 
strategic programs and opportunities for EPSCoR participants 
that stimulates sustainable improvements in their R&D capacity 
and competitiveness, and to advance science and engineering 
capabilities in EPSCoR jurisdictions for discovery, innovation 
and overall knowledge-based prosperity. Twenty-seven states 
plus Puerto Rico and the U.S. Virgin Islands are currently 
eligible for NSF EPSCoR support. These 29 jurisdictions 
comprise 20 percent of the U.S. population, 25 percent of the 
research in doctoral universities, and 18 percent of our 
nation's scientists and engineers. NSF EPSCoR funding is 
awarded through a merit-based peer review process.
    EPSCoR has been very beneficial to South Carolina. The 
Medical University of South Carolina has made relatively modest 
contributions to the creation of knowledge in science and 
engineering disciplines. However, with the assistance of 
programs like NSF EPSCoR, we are now poised to contribute in a 
substantial and sustainable way to the competitiveness of our 
Nation.
    The current NSF EPSCoR RII grant was awarded to South 
Carolina in July 2009. This RII has presented us with an 
exciting opportunity for South Carolina to implement a 
statewide vision for building a competitive edge in the 
emerging field of organ printing that can create human organs 
such as hearts, kidneys and blood vessels. This has ample depth 
and breadth to bring together faculty and students from nearly 
all of South Carolina's institutions of higher education to 
work together toward a common purpose. Furthermore, NSF EPSCoR 
funds have been leveraged through the recruitment of new 
professors to the State through the South Carolina Centers of 
Economic Excellence Act and the Research Universities 
Infrastructure Act.
    The NSF RII award provided the impetus for South Carolina 
and Tennessee to partner on a new NSF EPSCoR 
cyberinfrastructure award that provides personnel and equipment 
to facilitate coordination with Clemson's High Performance 
Computing support staff and with TeraGrid specialists.
    Finally, NASA EPSCoR funds have catalyzed connections among 
South Carolina's researchers and the NASA Jet Propulsion 
Laboratory to design and test a useful and efficient lunar 
wheel for use on a small pressurized rover that will enable 
astronauts to explore the moon.
    We believe that the EPSCoR program can be improved. 
Targeted options continue to be the most viable and effective 
pathways to develop the scientific infrastructure, talent and 
critical mass in the EPSCoR states. There should be continued 
investment in these competitive grant opportunities for states 
that meet EPSCoR criteria. The current EPSCoR program could be 
improved by dividing it into research, education and workforce 
components.
    With regard to the entire EPSCoR program, I would suggest 
that NSF set a goal of doubling the percentage of its funds 
annually that are awarded to the 27 EPSCoR states and two 
jurisdictions and slightly less than ten percent of NSF's 
annual R&RA obligations to 20 percent within ten years. We also 
need assurances that as the new states are added, the funding 
needed for them will be requested and appropriated.
    EPSCoR states have trained a lot of scientists and 
engineers over the years, and we need incentives to keep and 
bring new talent to our state. Physical infrastructure 
initiatives outside of the EPSCoR program could also be very 
useful. Cutting-edge facilities, renovations and equipment 
remain a major obstacle to competitiveness for the EPSCoR 
states. Our institution has over $100 worth of deferred 
maintenance in our research facilities. A separate program or a 
set-aside in existing programs would be very helpful.
    Finally, while South Carolina has made impressive progress 
in cyberinfrastructure, it has not been easy or inexpensive. 
Many of the EPSCoR states have not been as fortunate as South 
Carolina and are still lacking the bandwidth systems that will 
enable the modeling and computer simulations needed for climate 
change, biomedical and advanced research for visualization.
    In closing, we believe the value, effectiveness and 
sustainability of EPSCoR programs is very clear, both as a 
catalyst for improving our respective states, and to enhance 
America's overall competitiveness in the global economy. Again, 
I thank you for the opportunity to testify today.
    [The prepared statement of Dr. Raymond follows:]

                 Prepared Statement of John R. Raymond

    Mr. Chairman and Members of the Subcommittee, my name is Dr. John 
Raymond. I am Vice President for Academic Affairs and Provost at the 
Medical University of South Carolina. I have also served as Chair of 
the State of South Carolina Experimental Program to Stimulate 
Competitive Research (EPSCoR) Committee for the past eight years. Thank 
you for the opportunity to testify today regarding the research 
infrastructure needs of our universities and colleges including 
research facilities and cyber-infrastructure capability, the capacity 
of the research infrastructure to meet the current and future needs of 
U.S. scientists and engineers, and the appropriate role of the Federal 
government in sustaining such infrastructure.
    In this testimony, I have been asked to answer questions related to 
the current National Science Foundation EPSCoR grant awarded to South 
Carolina. Specifically, I will address EPSCoR's role in facilitating 
partnerships with state and local governments and the private sector to 
improve our research infrastructure, its leveraging effect on improving 
cyber-infrastructure capabilities, and its impact on the Medical 
University of South Carolina. Secondly, I will describe the state of 
research infrastructure and research facilities at the Medical 
University of South Carolina and our unmet research infrastructure 
needs. Thirdly, I will provide recommendations on how to improve the 
EPSCoR program based on the findings and recommendations of the EPSCoR 
Foundation.
    Before answering the three specific questions posed to me, it might 
be useful to provide a brief summary of the EPSCoR program and my 
university to place my answers into the appropriate context. The 
National Science Foundation EPSCoR program has a statutory function 
``to strengthen research and education in science and engineering 
throughout the United States and to avoid undue concentration of such 
research and education.'' This is accomplished through two goals, which 
are (1) to provide strategic programs and opportunities for EPSCoR 
participants that stimulate sustainable improvements in their R&D 
capacity and competitiveness; and (2) to advance science and 
engineering capabilities in EPSCoR jurisdictions for discovery, 
innovation and overall knowledge-based prosperity. South Carolina is 
one of the original NSF EPSCoR-eligible states designated in 1980 
(please see Figure 1). Twenty-nine jurisdictions including twenty-seven 
states, the Commonwealth of Puerto Rico, and the U.S. Virgin Islands 
are currently eligible to compete for support through various NSF 
EPSCoR mechanisms.\1\ Those 29 jurisdictions comprise 20 percent of the 
U.S. population, 25 percent of the research and doctoral universities, 
and 18 percent of the nation's scientists and engineers. NSF EPSCoR 
funding is awarded through a rigorous process of merit-based peer-
review to ensure quality, accountability and sustainability. Many other 
federal agencies support programs similar to the NSF EPSCoR program; 
for example, the National Institutes of Health has a program called the 
Institutional Development Award (IDeA) program.
---------------------------------------------------------------------------
    \1\ Eligible EPSCoR jurisdictions: Alabama, Alaska, Arkansas, 
Delaware, Hawaii, Idaho, Iowa, Kansas, Kentucky, Louisiana, Maine, 
Mississippi, Montana, Nebraska, Nevada, New Hampshire, New Mexico, 
North Dakota, Oklahoma, Puerto Rico, Rhode Island, South Carolina, 
South Dakota, Tennessee, U.S. Virgin Islands, Utah, Vermont, West 
Virginia, and Wyoming.



    Founded in 1824, the Medical University of South Carolina is a 
freestanding academic health science center composed of six health-
related colleges (Dental Medicine, Graduate Studies, Health 
Professions, Medicine, Nursing, Pharmacy). Until recently, our 
institution made relatively modest contributions to the creation of 
knowledge in science and engineering disciplines; with the assistance 
of programs like NSF EPSCoR, we now are poised to contribute in a 
substantial and sustainable way to the competitiveness of our nation. 
We were awarded extramural research funding of nearly $218 million in 
FY 2009-10, of which $140 million was from federal sources, and $103 
million from the National Institutes of Health.
    The current NSF EPSCoR Research Infrastructure Improvement (RIO 
grant was awarded to South Carolina in July 2009. This RII has 
presented an exciting opportunity for South Carolina to implement a 
statewide vision towards building a competitive edge in the emerging 
field of ``organ printing''--operationally defined as computer-aided, 
layer-by-layer deposition of biologically relevant material with the 
purpose of engineering functional tissues and organs. The idea is that 
we can use cultured cells and supporting materials as ``ink'' that can 
be built up using modified ink jet printers and powerful computers to 
create human organs such as hearts, kidneys, and blood vessels. The 
patient's own cells (such as fat cells) can be used to make these 
organs to provide a ready source for transplantation to treat and cure 
diabetes, kidney failure, heart failure and atherosclerosis. What 
patient with diabetes wouldn't donate some of their excess fat cells to 
make a new pancreas to cure their diabetes?
    Organ printing poses a grand challenge in terms of engineering and 
biological principles, and a grand opportunity for South Carolina to 
contribute to the competitiveness of our country. Currently, the 
thickness of printed tissue constructs is limited to four cell layers 
or less due to lack of a blood supply. In order to manufacture more 
complex organs, one must successfully engineer a vascular supply, which 
will require a 3-D tree-like network of blood vessels.
    The grand vision of this RII has ample depth and breadth to bring 
together faculty and students from nearly all of South Carolina's 
institutions of higher education to work toward a common purpose. The 
2009 SC NSF EPSCoR RII focuses on a diverse subset of institutions 
including three research intensive institutions (Clemson University, 
Medical University of South Carolina, University of South Carolina), 
three historically black colleges (Claflin University, South Carolina 
State University, Voorhees College), two other predominately 
undergraduate institutions (Furman University, USC-Beaufort) and 3 
technical colleges (Denmark Technical College, Greenville Technical 
College, York Technical College). Together we form the SC Alliance for 
Tissue Biofabrication.
    EPSCoR funds were essential for demonstrating the feasibility of 
using existing rapid prototyping equipment to print an intra-organ 
vascular tree. Drs. Vladimir Mironov and Roger Markwald at MUSC 
facilitated the fabrication of a 3-D ``plastic'' kidney (see Figure 2), 
which was recently printed based on a computer-aided design provided by 
Prof Nicolas Smith from the University of Oxford (UK) using expertise 
and facilities at 3-D Systems/York Technical College. This initial 
success and preliminary data strongly suggest that existing rapid 
prototyping technology using layer-by-layer addition of building blocks 
has sufficient resolution for bioprinting a complex branched vascular 
tree. Rapid prototyping is a rapidly growing, $100 billion/yr industry 
and 3-D Systems, Inc, located in Rock Hill, SC, is a leading global 
provider of 3-D printing, rapid prototyping and additive manufacturing 
products. This is an excellent example of EPSCoR funds being used to 
catalyze academic-industrial collaborations towards building an 
advanced biomanufacturing industry in South Carolina.



    The NSF EPSCoR funds have been leveraged through the recruitment of 
new professors to the state of South Carolina through the Centers of 
Economic Excellence Act, and the Research Universities Infrastructure 
Act, two key economic development initiatives passed by the South 
Carolina Legislature in 2002 and 2004, respectively. Those acts provide 
state matching funds for recruitment of endowed professors, and for 
research construction. We have used state funds and private sector 
matching funds to create multi-institutional Centers of Economic 
Excellence in Regenerative Medicine, and in Tissue Biofabrication. 
Several of the professors recruited to these centers have faculty 
appointments at Clemson, USC and MUSC, thus serving as bridges between 
our institutions. These new centers will be based in a new 100,000 
ft2 Bioengineering Building, which will be completed in late 
2011. This building will house engineers from Clemson and USC, and life 
scientists from MUSC, working in interdisciplinary teams to address 
grand challenges like the organ bioprinting project. We also have 
leveraged the NSF EPSCoR award by developing interdisciplinary 
educational programs that bring together students and faculty from the 
technical colleges, historically black serving institutions, four-year 
and research-intensive institutions.
    Finally, the NSF RII award provided the impetus for South Carolina 
and Tennessee to partner on a new NSF EPSCoR cyberinfrastructure award 
that provides personnel and equipment to facilitate coordination with 
Clemson High Performance Computing support staff and TeraGrid 
specialists. This cyberinfrastructure grant also enables South Carolina 
institutions to have access to the TeraGrid Kracken system housed at 
Oak Ridge National Laboratory. This grant, along with a $21 million 
award from The Duke Endowment and an $8 million award from the Federal 
Communication Commission, has allowed us to develop a high-speed, high-
bandwidth optical and wireless communication grid that spans the state 
and facilitates competitiveness.
    NASA EPSCoR funds have catalyzed connections among Dr. Joshua 
Summers' team at Clemson, and Michelin, Milliken and the NASA Jet 
Propulsion Laboratory to design and test a useful and efficient lunar 
wheel for use on the Small Pressurized Rover that will enable 
astronauts to explore the moon. The futuristic rover with its 
``tweels'' joined NASA astronauts in President Obama's inaugural parade 
on Pennsylvania Avenue. The accompanying Figure 3 shows Dr. Summers and 
undergraduate student Ms. Samantha Thoe inspecting the metallic 
prototype.



    Other federal agency EPSCoR funds have been applied to the areas of 
energy and alternative fuels. For example, Dr. Terry Tritt's research 
group at Clemson University has extensive interactions with Oak Ridge 
National Lab and Savannah River National Lab through the DOE EPSCoR 
Partnership Program. Dr. Tritt has received international attention for 
his study of thermoelectric energy, and on materials that can recapture 
``lost'' energy from ``wasted'' heat.
    These are just a few examples of how EPSCoR funds have been used to 
advance research and science education in South Carolina.
    With regard to MUSC's research infrastructure, we have a number of 
new, state-of-the-art research buildings focusing on childhood 
diseases, bioengineering and drug discovery and development. We also 
have a number of aging buildings that will require significant upgrades 
and renovations to accommodate our expanded scope of research; and new 
high-end instrumentation to enable our teams to perform the mass 
spectroscopy, magnetic resonance imaging, high capacity computing, 
emerging microscopic methods, and interactive teaching, materials 
sciences, and biofabrication, as well as other emerging methods. We 
share these needs with many educational institutions, even those in the 
research powerhouse states. The continued support of EPSCoR programs 
will be essential for our state, and for institutions like MUSC, to 
make sustainable contributions to scientific discovery, contemporary 
science and engineering, education, innovation and the overall 
competitiveness of our country.
    We believe targeted options continue to be the most viable and 
effective pathways to develop the scientific infrastructure, talent and 
critical mass in the EPSCoR states. There should be a continued 
investment in competitive grant opportunities for states meeting EPSCoR 
criteria. We believe the current EPSCoR program could be improved by 
dividing it into several components--(1) research and (2) education and 
workforce: Alternatively, we could simply adopt the NIH dual model of 
COBREs which are research center development grants, and INBREs which 
are state network grants to educate and train the next generation of 
biomedical scientists. This would be a much more direct approach to 
meeting both research infrastructure and ``pipeline'' needs. Each 
component should, of course, be adequately funded at levels similar to 
those at NIH.
    We would appreciate renewed efforts to involve EPSCoR states in the 
regular NSF programs. This means more representatives from EPSCoR 
states on the National Science Board, NSF Advisory committees and other 
relevant ``planning'' entities; more co-funding especially as the NSF 
budget is growing, and greater use of mechanisms that will ensure 
EPSCoR participation in major NSF initiatives. I believe that a few 
years ago, extra points were awarded for including EPSCoR states in 
certain applications for large programs. This should be reinstated. 
Other efforts should be made to assist EPSCoR states in participating 
in more large-scale NSF efforts such as Science and Technology Centers 
(STCs), Engineering Research Centers (ERCs), and Materials Research 
Science and Engineering Centers (MRSECs). Unless that is done, the 
dollar imbalance between the established states and the EPSCoR states 
will continue to grow. In this regard, I would suggest that NSF set a 
goal of doubling the percentage of its funds, annually, that are 
awarded to the 27 EPSCoR states and 2 jurisdictions--from slightly less 
than 10% to 20% within ten years. Then, coalesce some of the 
initiatives recommended above, as well as others gleaned from the 
broader EPSCoR community, into a ``Strategic Implementation Plan'' to 
meet that goal.
    We also need assurance that as new states are added, the funding 
needed for them is requested and appropriated. It costs $5-10 million a 
year to bring a new state into the EPSCoR program during its first five 
years and these new EPSCoR states tend to be more competitive than some 
of the existing ones. Consequently, it is self-defeating to drain 
resources from one to help the other.
    We should look at other mechanisms as well. EPSCoR states have 
trained a lot of scientists and engineers over the years who, 
regrettably, have then simply moved to other states. More are staying 
in our states as we build our infrastructure and attract innovative 
companies. We need incentives to keep and bring new talent to our 
states. Physical infrastructure initiatives outside of the EPSCoR 
program could also be useful. Renovations and' equipment remains a 
major obstacle to competitiveness for the EPSCoR states. Cutting edge 
facilities, renovations and equipment remain a major obstacle to 
competitiveness for the EPSCoR states. A separate program or a set 
aside in existing programs would be helpful.
    Physical infrastructure initiatives outside of EPSCoR or in 
addition to the existing EPSCoR program are essential. That is your 
focus today. The EPSCoR states unquestionably and unequivocally require 
such investments. Construction of scientific facilities, renovations 
and equipment remain a major obstacle to competitiveness in the EPSCoR 
states.
    Finally, while South Carolina has made impressive progress in 
cyberinfrastructure, it has not been easy or inexpensive. Many of the 
EPSCoR states have not been as fortunate and many are still lacking the 
bandwidth and support systems that will enable modeling and simulations 
needed for climate change, biomedical and advanced research and for 
visualization.
    We thank this subcommittee for its ongoing support of our states 
and for the wisdom to invest in programs that engage the populace of 
all of our states in building science and engineering capabilities that 
will broaden the base of talent and the capacity for innovation 
throughout the United States. We believe in the value, effectiveness 
and sustainability of EPSCoR programs--both as a catalyst for improving 
our respective states and to enhance America's competitiveness in the 
global economy.
    In closing, I thank you for the opportunity to address the 
Subcommittee today.

                     Biography for John R. Raymond




    Chairman Lipinski. Thank you, Dr. Raymond.
    Now I will recognize Dr. Dunning.

STATEMENT OF DR. THOM H. DUNNING, JR., DIRECTOR OF THE NATIONAL 
CENTER FOR SUPERCOMPUTING APPLICATIONS, UNIVERSITY OF ILLINOIS 
                      AT URBANA-CHAMPAIGN

    Dr. Dunning. Mr. Chairman, thank you for the opportunity to 
address the Subcommittee on Research and Science Education.
    Before I start, I would like to take this opportunity to 
thank Representative Ehlers for his service to the Nation's 
science and educational enterprise. We all have greatly 
benefited from your dedication to advancing science and 
engineering, and we thank you.
    Now let me return to the topic at hand, the state of 
cyberinfrastructure in the United States. To ensure that we are 
all on the same page, I would like to note that 
cyberinfrastructure consists of computing systems, and consists 
of data sources, data storage systems, visualization 
environments all linked by high-speed networks, knitted 
together by software and enabled by expert support staff. 
Cyberinfrastructure allows us to make discoveries and 
innovations not otherwise possible, and as such, it is now a 
critical part of the Nation's research infrastructure.
    Given the time constraints, I will focus my comments mainly 
on the high-performance computing, or HPC, aspect of 
cyberinfrastructure including the University of Illinois's Blue 
Waters, which, when it comes online in 2011, will be the most 
powerful computer in the Nation for open scientific research, 
and in fact, it will likely be the most powerful computer in 
the world for such research. This extraordinary computer will 
be capable of sustaining a million billion arithmetic 
operations per second, have more than one petabyte of memory, a 
million times what you have in your PC, more than 10 petabytes 
of online disc and 500 petabytes, or half an exabyte, of 
archival storage.
    National resources like Blue Waters are not the result of 
one organization's work. This computer reflects a model, in 
fact, of a university-state-federal-industry partnership 
founded on a 24-year relationship between the University of 
Illinois at Urbana-Champaign, the State of Illinois and the 
National Science Foundation. In the specific case of Blue 
Waters, the State of Illinois built the national petascale 
computing facility and provided I-wire connectivity to connect 
that facility to the national research networks. The University 
of Illinois is buying the archive system and networking gear 
and investing in software for Blue Waters. NSF is buying the 
computing system and funding its operation and maintenance. And 
finally, IBM, which is the computer vendor for Blue Waters, is 
working closely with the university to ensure that Blue Waters 
delivers maximum value to the scientific community.
    In this regard, I do have one concern. Continuing progress 
requires that industry be researching, developing and producing 
high-end computers for scientific research, and I am very 
concerned at the drop-off in companies investing in this 
particular area, especially in the highest end.
    Let me now make four comments on the status of 
cyberinfrastructure for high-performance computing and what is 
needed for this cyberinfrastructure to be effectively and 
efficiently used for discovery and innovation. First, let me 
note that the deployment of fast high-end computing systems by 
NSF has been very successful, providing extraordinary value to 
the scientific community, but what has been lacking is the 
investment in user support. Even scientists experienced with 
high-performance computing require assistance to use systems at 
the leading edge, and as we have found at NSCA and other sites, 
researchers in a growing number of other fields are finding 
high-performance computing critical to meeting their particular 
goals. Expert support is required to bring them into the fold.
    A second concern about NSF's HPC program is the frequency 
of competitions associated with the deployment of these 
resources. Competition is good, but when you are building 
infrastructure, completeness, robustness and continuity are 
also critical. Too-frequent competitions make it difficult to 
attract high-quality staff, result in discontinuities and 
inefficiencies in support service and are a drain on valuable 
staff support time. To be blunt, the current model is 
unsustainable and a task force currently advising NSF on future 
HPC strategies will recommend longer term, more stable funding, 
coupled with rigorous reviews to ensure quality.
    A third concern is the balance between investments in 
hardware and software. Scientists and engineers certainly need 
access to ever more powerful computers, but science and 
engineering applications must be carefully designed to fully 
exploit the capabilities of the high-performance computing 
systems available. New tools and approaches are needed to help 
scientists develop applications for Blue Waters and the even 
bigger computers that will come next. This is a major area for 
NSF investment and will require significant collaboration 
between NSF directorates and offices on developing a new 
generation of science and engineering applications as well as a 
robust and complete HPC software stack.
    A fourth area of concern, not just for HPC but also for 
cyberinfrastructure in general, is networking. While the United 
States may appear to be in good shape on the surface, this 
smooth surface hides a number of issues. Scientists are 
choosing not to undertake some activities because they know 
those activities will stress the networks. Data volumes are 
rapidly increasing and will overwhelm current capacities in the 
next few years.
    And finally, as mentioned by Dr. Raymond, networking 
capabilities are not evenly distributed. Many universities may 
not be able to benefit from the major advances being made in 
data-intensive science.
    To tackle some of these current and upcoming challenges in 
networking, NSF must do more. One key area of need is to 
interconnect NSF's major research facilities, instruments and 
computers. A potential model for this is the Department of 
Energy's ESNET, which is a high-performance network being built 
to connect major Office of Science research facilities. Another 
key area, and one that is specific to NSF, is the need to 
enhance the ability of university researchers to connect to 
NSF's major facilities, the so-called end-to-end problem. All 
of the concerns I have raised are driven by the need to tackle 
not only existing problems, but to prepare for future 
opportunities such as the coming revolution in data-driven 
discovery.
    One of the most exciting advances in science and 
engineering is the increasing digitization of observational 
science, from astronomy to biology to environmental science. 
Advanced sensor arrays, microscopes and automated sequencers, 
and telescopes are allowing us to produce huge quantities of 
meaningful data. At NSF, this can clearly be seen in its large 
MREFC projects. To build the cyberinfrastructure for these 
projects, we need to share and reutilize software, whenever 
possible, that is both costly to build and maintain. Such 
coordination is not easy with independent projects, and the 
lack of continuity at supercomputing centers leads to their 
under utilization by these major data-driven discovery 
projects.
    I would like to conclude with a brief word about education 
and cyberinfrastructure. Two key questions that we have are: 
what does the next generation of scientists and engineers need 
to know about cyberinfrastructure, and second, how can we 
modify the curriculum and courses to provide the needed 
knowledge? NSF with its broad mandate in science and 
engineering, research and education is well suited to explore 
the options and serve as a catalyst for the needed changes at 
universities.
    This concludes my verbal remarks. Thank you for the 
opportunity to testify before you today, and I am more than 
happy to answer any questions you may have.
    [The prepared statement of Dr. Dunning follows:]

               Prepared Statement of Thom H. Dunning, Jr.

What Is Cyberinfrastructure?

    Cyberinfrastructure, n., cyberinfrastructure consists of computing 
systems, data sources and data storage systems, visualization 
environments, and support staff, all linked by high speed networks to 
make discoveries and innovations not otherwise possible.
    Over the past quarter century, computing has become an integral 
part of the fabric of experimental and theoretical science. All but the 
simplest laboratory experiments are performed under computer control, 
the data is analyzed using software running on a personal computer or 
small compute cluster, and the results compared with the latest 
theories through computational simulations on high performance 
computers. The use of computing technology is now spreading to the 
observational sciences, which are being revolutionized by the advent of 
powerful new sensors that can detect and record a wide range of 
physical, chemical and biological phenomena-from massive digital 
detectors in a new generation of telescopes to sensor arrays for 
characterizing ecological and geological areas and new advanced 
sequencing instruments for genomics research.




Research Advances Enabled by Cyberinfrastructure

    Three major modes of scientific discovery are enabled by 
cyberinfrastructure: computational modeling and simulation, data-driven 
discovery, and, increasingly, the coupling of these two modes. To 
address the questions posed by the Subcommittee, I will discuss the 
cyberinfrastructure needs of these three modes of scientific discovery 
and then provide an analysis of the status of the existing 
cyberinfrastructure. To begin, let us briefly review the science and 
engineering advances made possible by cyberinfrastructure.
    Computational Modeling and Simulation. In computational modeling 
and simulation, scientists develop a mathematical model of the 
phenomena of interest, e.g., the chemical and physical processes 
involved in an internal combustion engine or the processes involved in 
the prediction of weather, and then use high performance computers to 
solve the resulting equations. For most phenomena of interest, the 
equations are very complex and, so, the power of computational modeling 
and simulation grows with increases in computing power. As computing 
systems have progressed from the megaflops era in the 1970s to the 
petaflops era of today, our ability to accurately simulate a broad 
range of biological, chemical, physical and, even, social phenomena has 
grown dramatically.

          The Southern California Earthquake Center seeks to 
        develop a predictive understanding of earthquake processes 
        aimed at providing society with improved understanding of 
        seismic hazards. In partnership with earthquake engineers, SCEC 
        researchers are developing the ability to conduct end-to-end 
        simulations [``rupture to rafters'') to extend this improved 
        understanding of seismic hazards to an improved understanding 
        of earthquake risks and risk mitigation strategies.

          Researchers at the University of Illinois at Urbana-
        Champaign are using computational simulations to obtain a 
        detailed understanding of the functioning of the ribosome, the 
        large cellular machine responsible for synthesizing proteins in 
        our cells, as well understanding the mechanism used by the 
        poliovirus to gain entry into our cells. The former will 
        enhance our fundamental understanding of cell biology, while 
        the latter may lead to the development of better anti-viral 
        drugs.

          A team from Michigan State University and the 
        University of California, San Diego are studying the formation 
        of the first galaxies. Based on a fundamental understanding of 
        the physical processes and the initial conditions that led to 
        the formation of the first stars, powerful numerical 
        simulations are helping astrophysicists understand how and when 
        the very first sources of light formed.

    All of these simulations are numerical- and data-intensive and can 
only be performed on the most powerful computers available.
    Data-driven Discovery. In data-driven discovery, scientists gather 
information from various data sources, e.g., a large digitally-enabled 
telescope, an array of environmental sensors, or ``gangs'' of genome 
sequencers, and then analyze the resulting mass of data 
usingsophisticated mathematical procedures seeking patterns, 
information and understanding. Data-driven discovery requires an 
extensive cyberinfrastructure that supports data collection and 
transport to storage sites, followed by data cataloging, integration 
and analysis (including visualization). Often, the cataloged data 
becomes a resource for a large research community. Depending on the 
quantities of data involved as well as the mathematical demands of the 
analysis, data-driven discovery may require extensive computing 
resources as well as large data storage facilities.

          The Ocean Observatory Initiative is constructing an 
        integrated observatory network to provide the oceanographic 
        research and education community with: (i) a cabled network of 
        monitoring devices on the sea floor spanning important 
        geological and oceanographic features, (ii) an array of 
        relocatable deep-sea buoys that can be deployed in harsh 
        environments, and (iii) construction of new facilities or 
        enhancements to existing facilities leading to an expanded 
        network of coastal observatories. The OOI will provide earth 
        and ocean scientists with unique opportunities to study 
        multiple, interrelated processes over timescales ranging from 
        seconds to decades; to conduct comparative studies of regional 
        processes and spatial characteristics; and to map whole-earth 
        and basin scale structures.\1\
---------------------------------------------------------------------------
    \1\ See: http://www.oceanleadership.org/programs-and-partnerships/
ocean-observing/.

          The Large Synoptic Survey Telescope (LSST) is unlike 
        other ground-based telescopes. It is a wide-field survey 
        telescope and camera that can image the entire sky in just 
        three nights, providing a time history of celestial events. 
        Using an 8.4-meter ground-based telescope, the LSST will, for 
        the first time, produce a wide-field astronomical survey of our 
        universe. Its 3 gigapixel camera--the world's largest digital 
        camera--will provide digital imaging of faint astronomical 
        objects. The LSST will provide unprecedented three-dimensional 
        maps of the mass distribution in the universe, in addition to 
        the traditional images of luminous stars and galaxies, These 
        maps will be used to better understand the nature of the 
        mysterious dark energy that is driving the accelerating 
        expansion of the universe. In addition, the LSST will also 
        provide a comprehensive census of our solar system, including 
---------------------------------------------------------------------------
        potentially hazardous near-Earth asteroids

    Data-driven Computational Modeling and Simulation. There are 
increasing opportunities for linking data-driven discovery with 
computational modeling and simulation. For example, in the NSF-funded 
LEAD project (Linked Environments for Atmospheric Discovery),\2\ one of 
the goals is to gather and analyze the data from a distributed array of 
Doppler radars to determine, in real time, when atmospheric conditions 
are ripe for the formation of a tornado and then launch computational 
simulations to determine the likely path and intensity of the tornado. 
Such opportunities will grow in the future as sources of sensed data 
become more widespread.
---------------------------------------------------------------------------
    \2\ See: https://portal.leadproject.org/gridsphere/gridsphere.

---------------------------------------------------------------------------
Development of a National Cyberinfrastructure

    In recognition of the increasing importance of research 
cyberinfrastructure, the National Science Foundation recently issued a 
Dear Colleague Letter on ``Cyberinfrastructure Framework for 21st 
Century Science and Engineering.'' This letter stated that it was 
imperative for NSF to develop a long term vision for the nation's 
cyberinfrastructure that covered the following critical areas:

        1.  Cyberinfrastructure for:

                a.  High end computational, data, visualization and 
                sensor-based systems and the associated user support 
                for transformative science.

                b.  NSF's large-scale collaborative research facilities 
                and projects, integrated with that of other federal 
                agencies and international organizations.

        2.  Linkage of this cyberinfrastructure into campuses 
        (including government and businesses) accompanied by programs 
        that support integrated, widely dispersed, broadly based 
        activities and resources.

        3.  Education and outreach to help develop computational 
        science- and technology-savvy researchers and workforce.

    This letter was signed by all of the Assistant Directors at NSF as 
well as the directors of many major NSF programs.
    The development of a national cyberinfrastructure for research 
poses many unique challenges for NSF. Cyberinfrastructure is very 
different from physical infrastructure such as a laboratory building. 
Computing and related technologies are still rapidly advancing--
computing power doubles every two years, disk capacity increases even 
more rapidly, 60% per year. The software that ties all of the 
infrastructure elements together to create a unique research capability 
has to be revised in response to these changes in technology. Finally, 
the use of cyberinfrastructure is still in its infancy--high quality 
support staff are needed to ensure that the U.S. research community can 
take advantage of the new capabilities provided by cyberinfrastructure. 
This close coupling of hardware, software, and expertise with a rapidly 
changing technology base is unparalleled in other types of 
infrastructure.
    Cyberinfrastructure must also be funded through different 
mechanisms. Infrastructure must be sustained over long periods of time 
to be useful to researchers, and it cannot be sustained through a 
series of short term, loosely integrated projects. Like an interstate 
highway, cyberinfrastructure must provide a smooth, continuous path 
from one point to another. On the other hand, cyberinfrastructure must 
also evolve as computing technology advances; otherwise, it will 
rapidly become outdated. So, there must be flexibility in how the 
funding is used in long term cyberinfrastructure projects. Finally, 
cyberinfrastructure is expensive, both in terms of the hardware that 
must be deployed, the software that must be developed and maintained, 
and the support staff that are critical for its efficient functioning. 
It is important to avoid duplication and leverage existing capabilities 
and resources whenever possible.
    The NSF-wide Advisory Committee for Cyberinfrastructure has begun 
work on the development of the new cyberinfrastructure framework 
outlined in the Dear Colleague letter,\3\ establishing six Task Forces:
---------------------------------------------------------------------------
    \3\ See: https://nsf.sharepointspace.com/acci-public/
default.aspx.

---------------------------------------------------------------------------
Work Force Development

    The Task Forces involve distinguished scientists and engineers from 
across the nation as well as NSF program officers. Although the Task 
Forces are in the early stages of their work, they have already held a 
number of meetings and teleconferences to explore and discuss new 
concepts and strategies for developing a comprehensive national 
cyberinfrastructure. I am participating in three of these Task Forces: 
Grand Challenges, Software and Tools, and High Performance Computing 
and have colleagues who are involved in the other three Task Forces. 
This testimony provides a prologue to the work of the six NSF Task 
Forces.
    Before moving on, I should note that NSF is not the only federal 
agency that supports cyberinfrastructure in the nation's universities. 
The Office of Science in the U.S. Department of Energy (DOE-SC) also 
funds cyberinfrastructure for university researchers. DOE-SC has a well 
defined, long term plan to provide computational, data and 
communications resources for laboratory and academic researchers based 
on the identified needs of its major research programs. However, with 
the exception of the INCITE program,\4\ DOE-SC's cyberinfrastructure is 
closely tied to its mission needs, serving only those laboratories and 
universities deemed critical to that mission. The National Institutes 
of Health (NIH) supports a number of cyberinfrastructure-related 
software development efforts in its biomedical research programs but, 
by and large, depends on agencies such as NSF as well as the academic 
institutions that it supports to provide much of its 
cyberinfrastructure, especially the hardware. However, biomedical 
research is approaching a tipping point-the amount of data being 
accumulated in NIH's research programs will soon far exceed that which 
can be stored, managed and analyzed by the other agencies and 
institutions. NIH has several strategic planning activities underway to 
identify the best path forward. Whatever the outcome of these planning 
activities, meeting the growing computing and data needs of NIH's 
intramural and extramural research programs will surely require 
substantial increases in NIH's cyberinfrastructure investments.
---------------------------------------------------------------------------
    \4\ See: http://www.er.doe.gov/ascr/incite/index.html.

---------------------------------------------------------------------------
High Performance Computing

    As noted earlier, advances in computational modeling and simulation 
are driven by increases in computing power. Over the past few decades, 
increases in computing power have largely been driven by Moore's Law, 
with a doubling in computing power occurring every 18-24 months. Thus, 
the end of the 1980s saw the deployment of computers capable of 
performing a billion arithmetic operations per second.\5\ Ten years 
later, computing technology had advanced to the point that it was 
possible to perform a trillion arithmetic operations per second. In 
2008, computers capable of a quadrillion operations per second were 
deployed. It is expected that exascale computers, 1,000 times more 
powerful than petascale computers, will arrive in another eight years, 
although many hardware and software challenges must first be overcome.
---------------------------------------------------------------------------
    \5\ A typical arithmetic operation is the multiplication of two 14-
digit numbers to yield a 14-digit result.
---------------------------------------------------------------------------
    The National Science Foundation (NSF) and the Office of Science in 
the U.S. Department of Energy (DOE-SC) have committed to providing high 
performance computing resources for open scientific and engineering 
research, including for researchers who are funded by other government 
agencies. DOE-SC is funding several major computing centers in support 
of its energy and environmental missions as well as its broader 
national science mission: its flagship facility at Lawrence Berkeley 
National Laboratory and its leadership computing facilities at Oak 
Ridge National Laboratory and Argonne National Laboratory. NSF funds 
large national computing facilities at the Texas Advanced Computing 
Center and University of Tennessee/Oak Ridge National Laboratory and 
its largest national facility at the University of Illinois at Urbana-
Champaign. Although I am familiar with DOE's computing program, I will 
only discuss NSF's program here since NSF's programs are most relevant 
to the Subcommittee's charge. However, DOE-SC's contributions to the 
national cyberinfrastructure should be kept in mind.
    Cyberinfrastructure for High Performance Computing. NSF's high 
performance computing plan for 2006-10 was outlined in the document 
``Cyberinfrastructure Vision for 21st Century Discovery'' (March 2007). 
The report recognized the need, first articulated in the Branscomb 
report,\6\ for a broad range of computing resources, from leadership-
class national computing resources to university high performance 
computers and the compute/data clusters and workstations used by small 
research groups-the so-called Branscomb pyramid.\7\ The report stated 
NSF's intent to fund the highest performance computing systems, the so-
called Track 1 and Track 2 systems, as national resources. It 
envisioned that, in the 2006-10 time frame, Track 2 systems would 
provide 500+ teraflops (TF) of peak computing power and a Track 1 
system would provide a sustained performance approaching 1 petaflop 
(PF) on a broad range of science and engineering applications.\8\
---------------------------------------------------------------------------
    \6\ ``From Desktop to Teratlop: Exploiting the U.S. Lead in High 
Performance Computing,'' NSF Blue Ribbon Panel on High Performance 
Computing, Lewis Branscomb (chairman), NSF 93-205, August 1993.
    \7\ NSF supports the acquisition of computer systems at the lower 
levels of the Branscomb pyramid through many other programs, e.g., the 
Major Research Instrumentation (MRI) program. See: http://nsf.gov/pubs/
2010/nsf10529/nsf10529.htm.
    \8\ The peak performance of a computer system is the theoretical 
limit of its computing capability; it can never be achieved. The 
sustained performance of a computer is the performance that is actually 
achieved on a given science or engineering application. Although peak 
performance is used as a proxy for sustained performance, the 
correlation can be very weak.
---------------------------------------------------------------------------
    NSF awarded funding for Track 2 systems to the Texas Advanced 
Computing Center (TACC) in 2006 (Sun system with a peak performance of 
579 TF) and the University of Tennessee/Oak Ridge National Laboratory 
in 2007 (Cray system with a peak performance of 1,028 TF). NSF 
announced the award of a Track 2 system to Pittsburgh Supercomputing 
Center in 2008. Unfortunately, the downturn in the economy led to the 
demise of the selected computer vendor, Silicon Graphics, Inc., which 
was acquired by Rackable Systems. Rackable Systems subsequently changed 
its name to SGI but cancelled the on-going contract negotiations with 
PSC. So, a third Track 2 system has not been deployed, despite clear 
evidence of a need for additional computing resources in the national 
allocation process run by NSF.
    To complement the Track 2 systems, NSF has also deployed a number 
of experimental and specialized computing systems to serve the nation's 
scientists and engineers, These include the many-core system deployed 
at the University of Illinois at Urbana-Champaign and another under 
development at the Georgia Institute of Technology, the data system 
being deployed at the San Diego Supercomputing Center, the experimental 
grid test-bed system at Indiana University, and the visualization 
systems at the University of Tennessee/Oak Ridge National Laboratory 
and the Texas Advanced Computing Center.
    In August 2007, NSF announced that it had selected the University 
of Illinois at Urbana-Champaign and its National Center for 
Supercomputing Applications (NCSA), IBM Corporation, and the Great 
Lakes Consortium for Petascale Computation to develop and deploy the 
Track 1 system called Blue Waters \9\ by July 1, 2011. Blue Waters is 
based on the most advanced technologies under development at IBM. These 
technologies are embodied in PERCS (Productive, Easy-to-Use, Reliable 
Computing System), which IBM is developing with funding from DARPA's 
High Productivity Computing Systems (HPCS) program. Blue Waters will be 
the first production deployment of PERCS and will be a truly 
extraordinary resource for science and engineering research.
---------------------------------------------------------------------------
    \9\ See: http://www.ncsa.illinois.edu/BlueWaters.
---------------------------------------------------------------------------
    Blue Waters will have more than 300,000 compute cores, 1 petabyte 
of main memory, 10 petabytes of user disk storage, and 500 petabytes of 
archival storage. It will have an innovative low latency, high 
bandwidth communications network that will facilitate scaling to large 
numbers of compute cores, and an I/O subsystem that will enable the 
solution of the most challenging data-intensive problems. With a peak 
performance of approximately 10 petaflops, performance analyses 
indicate that Blue Waters will sustain 1 petaflops or more on a broad 
range of science and engineering applications.
    The breakthroughs enabled by the extraordinary computing 
capabilities of Blue Waters will revolutionize many areas of science 
and engineering. In the past two years, NSF has awarded allocations of 
time or provisional allocations of time to eighteen (18) research teams 
from thirty (30) institutions across the country, with more to follow 
in future years. Research to be carried out on Blue Waters covers all 
areas of science and engineering from astronomy through biology, 
chemistry and materials science to geosciences and social and 
behavioral sciences.
    The Blue Waters Project is based on a 24-year partnership between 
the state of Illinois, the University of Illinois at Urbana-Champaign, 
and the National Science Foundation. To ensure the success of the Blue 
Waters Project, the state of Illinois agreed to provide a new state-of-
the-art, energy efficient facility to house Blue Waters. In addition, 
the University of Illinois at Urbana-Champaign is making substantial 
investments in the development of software for Blue Waters--
collaborating with IBM and the Great Lakes Consortium to: (i) enhance 
the systems software for Blue Waters, (ii) develop software and tools 
to facilitate the development of science and engineering applications 
for Blue Waters, and (id) aid scientists and engineers in rewriting 
their applications to obtain maximum performance on Blue Waters. In 
addition, previous investments by the state of Illinois in I-WIRE,\10\ 
a high performance communications infrastructure connecting the major 
research universities and laboratories in Illinois, provides the 
transport mechanism for connecting Blue Waters to national research and 
education networks.
---------------------------------------------------------------------------
    \10\ See: http://www.iwire.org/.
---------------------------------------------------------------------------
    Status of High Performance Computing Cyberinfrastructure. I will 
discuss the status of computer hardware and software for high 
performance computing separately as the issues are distinct, if 
interconnected.
    Computer Hardware. NSF has been successful in deploying new 
computing systems that are delivering extraordinary value for the U.S. 
research community--the first system delivered to TACC exceeded the 
total computing capacity of NSF's TeraGrid by a factor of more than 
five. However, the focus of these acquisitions was on the delivery of 
raw computing cycles and the funding available to provide support for 
the users of these new high performance computer systems was limited, 
This is unfortunate because this approach favors those scientists and 
engineers who are already using supercomputers and need little 
assistance, while our experience at NCSA and that at many other centers 
indicates there is a growing need for high performance computing 
resources in almost all fields of science and engineering. Without 
adequate user support, it will be difficult for these new researchers 
to make effective use of the available resources. High quality support 
staff is one of the most valuable resources in NSF's supercomputing 
centers and a fully funded user support program is needed.
    Both the Track 1 and Track 2 awards were made through open 
competitions that included the existing centers as well as many new 
entrants. The outcome of these competitions is that the two Track 2 
awards went to new centers--the Texas Advanced Computing Center and 
University of Tennessee/Oak Ridge National Laboratory. This is not 
necessarily bad, although it represents a lost of significant 
capability at San Diego Supercomputer Center and Pittsburgh 
Supercomputing Center. At this point the long term impact of the loss 
of funding on SDSC and PSC is unknown, but the potential loss of 
expertise at these sites is of great concern to the computational 
science and engineering research community.
    It should also be noted that the prospect of continual competitions 
has a corrosive effect on the staff at the centers--it is not only 
difficult to hire quality staff with funding that only lasts for 4-5 
years, but enormous amounts of staff time have to be dedicated to 
preparing for the competitions, rather than assisting researchers. The 
advantages of competition must be carefully balanced against those of 
stability in NSF's supercomputing centers program.
    The above problems have been extensively discussed by the Task 
Force on High Performance Computing. It is clear that stability and 
sustainability are critical if NSF's supercomputing centers are to 
attract high quality staff who can advance the use of high performance 
computing across the frontiers of science and engineering. Increased 
stability of the supercomputing centers that NSF supports, coupled with 
a rigorous review process to ensure operational quality, will certainly 
be one of the major recommendations from the Task Force. For additional 
thoughts on this topic, see the published comments by Larry Smarr \11\ 
and Sid Karin,\12\ the founding directors of NCSA and SDSC, 
respectively.
---------------------------------------------------------------------------
    \11\ ``The Good, the Bad and the Ugly: Reflections on the NSF 
Supercomputer Center Program,'' Larry Smarr, HPCWire, January 4, 2010 
(http://www.hpcwire.com/features/The-Good-the-Bad-and-the-Ugly-
Reflections-on-the-NSF-Supercomputer-Center-Program-80658282.html).
    \12\ ``Thoughts, Observations, Beliefs & Opinions About the NSF 
Supercomputer Centers,'' Sidney Karin, HPCWire, January 28, 2010 
(http://www.hpcwire.com/features/Thoughts-Observations-Beliefs-
Opinions-About-the-NSF-Supercomputer-Centers-82972987.html).
---------------------------------------------------------------------------
    Computer Software. During my two years as Assistant Director for 
Scientific Simulation in DOE's Office of Science, I played a central 
role in crafting its Scientific Discovery through Advanced Computing 
(SciDAC) program. This program highlighted the intimate connection 
between hardware and software and sought to advance computational 
modeling and simulation through balanced investments in these two 
areas. Experiences from this program, as well as DOE's ASCI program 
clearly show that advancing our ability to model complex natural 
systems requires as much, if not more, investment in software than in 
hardware.
    This problem is actually more acute now than when the SciDAC 
program was initiated. Since 2004, because of rapidly increasing 
thermal loads, the speed of a single compute core has not increased. 
Instead, computer vendors are adding additional cores to the chips and 
running the chips at lower speeds (to reduce the heat load). As a 
result, most laptops now use at least dual-core chips and quad-core 
chips are found in large compute servers, with eight-core chips now 
available from Intel and IBM. This trend has two major impacts:

        1.  Science and Engineering Applications. In the future, 
        increases in the performance of computational modeling and 
        simulation codes will only be achieved through the use of 
        larger and larger number of processors. Although this 
        ``scalability'' problem has been with us for nearly twenty 
        years, for much of that time its impact was not felt because of 
        the dramatic increases in the performance of single cores--a 
        factor of two orders of magnitude from 1989 to 2004. With 
        single core performance now stalled, computational scientists 
        and engineers must confront the scalability problem head on.

            The need for ever more scalability has increased the 
        difficulty of developing science and engineering applications 
        for high performance computers. At the heart of the problem is 
        algorithms that scale well to large numbers of compute cores. 
        This problem can only be solved through inspired research. But, 
        even given the appropriate algorithms developing science and 
        engineering applications for computers with hundreds of 
        thousands of compute cores, hundreds of terabytes of memory and 
        tens of petabytes of disk storage is challenging, The software 
        must be written, debugged, optimized and, to the extent 
        possible, made resilient to computer faults (e.g., the loss of 
        a compute core)--none of which is easy or straightforward. 
        Progress will require the creation of new software development 
        tools or the revision of existing tools (compilers, debuggers, 
        libraries, performance analysis tools, etc.) and integration of 
        these tools into a robust, easy-to-use application development 
        environment.

        2.  Computing System Software. Although computer companies 
        provide the base computing system software for high performance 
        computers, enhancements to this base software can greatly 
        facilitate operation, control and use of the system. This 
        problem is becoming more acute as the computer systems become 
        larger and more complex. Recently, a large international group 
        of computer and computational scientists has come together to 
        discuss plans for the development of software for petascale and 
        exascale computers.\13\ They are exploring how laboratories, 
        universities, and vendors can work together to coordinate the 
        development of a robust, full featured software stack for 
        petascale and -beyond computers.
---------------------------------------------------------------------------
    \13\ See http://www.exascale.org/iesp/Main-Page.

    The development of high performance computing software--science and 
engineering applications and computing systems software--is a topic 
that is being heavily discussed in several NSF Task Forces (Grand 
Challenges, Software and Tools, High Performance Computing, and Data). 
What is clear is that the current approach to developing a high 
performance computing software stack is too fragmented. The Task Forces 
have noted the need for long term, multi-level efforts in high 
performance computing software that involves all of NSF's directorates 
and the Office of Cyberinfrastructure. A partnership between OCI and 
the Computer & Information Science & Engineering directorate would help 
create software to manage, control and operate petascale and beyond 
computers as well as the new tools and software development environment 
needed to develop science and engineering applications for these 
computers. A partnership between OCI and the other directorates at NSF 
would foster the development of a new generation of science and 
engineering applications that can take full advantage of the power of 
petascale and beyond computers and realize the promise of these 
extraordinary resources for advancing science and engineering. Such 
partnerships already exist, e.g., the Accelerating Discovery in Science 
and Engineering through Petascale Simulations and Analysis (PetaApps, 
NSF 08-592) program, but they could be substantially strengthened.
    High Performance Computer Vendors. There is one last concern that 
deserves to be mentioned--the dwindling number of supercomputer vendors 
in the U.S. just a few years ago, five companies were involved in the 
development and deployment of supercomputers: IBM, Cray, Sun, SGI and 
Hewlett-Packard. Sun has now been subsumed by Oracle and SGI has been 
taken over by Rackable Systems. Although the long term consequences of 
these actions are not yet known, it is unlikely that Oracle and 
Rackable Systems/SGI will have as strong an interest in supercomputing 
as the original companies. Of the remaining companies, only IBM and 
Cray are actively involved in research and development on 
supercomputing. Although you would have to talk with these companies to 
better understand the issues surrounding this situation, it is clear 
than the supercomputing industry in the U.S. is not as healthy as it 
was just a few years ago.

Advanced Information Systems

    One of the most exciting research advances in science and 
engineering in the past decade is the digitization of observational 
science. Fields as disparate as astronomy, biology and environmental 
science are being revolutionized by the use of digital technologies: 
digital detectors like those in digital cameras in astronomy, highly 
automated sequencers in biology, and sensor arrays in environmental 
science. Data-driven discovery requires sophisticated, advanced 
information systems to collect, transport, store, manage, integrate and 
analyze these increasingly large amounts of invaluable data. The 
knowledge gained from data-driven discovery is already transforming our 
understanding of many natural phenomena and the future is full of 
promise.
    National Observatories. National astronomy observatories are major 
investments in the NSF research portfolio. At the leading edge of this 
portfolio are the latest additions to the NSF's list of approved major 
research equipment and facilities: the Atacama Large Millimeter Array 
\14\ (ALMA) and the Advanced Technology Solar Telescope \15\ (ATST). In 
addition, two other observatories are in the planning phases: the Giant 
Segmented Mirror Telescope \16\ (GSMT), which will operate in the 
ultraviolet to the mid-infrared with unprecedented resolution and 
sensitivity, and the Large-aperture Synoptic Survey Telescope \17\ 
(LSST), which will be able to image faint astronomical objects across 
the sky, including objects that change or move.
---------------------------------------------------------------------------
    \14\ See: http://www.almaobservatory.org/.
    \15\ See: http://atst.nso.edu/.
    \16\ See: http://www.gsmt.noao.edu/.
    \17\ See: http://www.lsst.org/lsst.
---------------------------------------------------------------------------
    NCSA is heavily involved in the LSST project and has been 
designated as the main storage and distribution site for all of the 
data produced by the telescope's 3.2 gigapixel camera. The data 
challenges to be faced by the LSST are typical of next generation 
optical telescopes, although the data-processing needs of the Square 
Kilometer Array (SKA) radio-telescope will dwarf those of the LSST. The 
LSST will produce more than 15 terabytes of data per night, yielding 
several petabytes of data per year, and 200 petabytes over its 
lifetime. This data rate, when combined with the need for real-time 
analysis to identify and characterize changing or moving objects as 
well as traditional data mining on petabyte-size data sets, requires a 
new approach to data management, automated processing, and analysis. 
Although the telescope will not see first light until 2014, NCSA is 
already working with other partners in the LSST project to design the 
cyberinfrastructure needed to meet these challenges.
    Several national-scale environmental observatories are also major 
initiatives in the current NSF research and development portfolio. 
These are represented by the Ocean Observatory Initiative \18\ (001), 
which is leading the way in this space, along with the National 
Ecological Observatory Network \19\ (NEON), and the WATERS Network.\20\ 
Ecological observatories have been in existence for many years with one 
of the oldest large-scale observatories being the Long-Term Ecological 
Research Network,\21\ although the grand challenges being addressed and 
the level of integration required for the new observatories far exceeds 
those of earlier observatories.
---------------------------------------------------------------------------
    \18\ See: http://www.oceanleadership.org/programs-and-partnerships/
ocean-observing/ooi/.
    \19\ See: http://www.neoninc.org/.
    \20\ See: http://www.watersnet.org/.
    \21\ See: http://www.lternetedu/.
---------------------------------------------------------------------------
    Environmental science often depends upon observations from multiple 
observatories not only of the same type but also complementary 
observatories. For instance researching the effects of climate change 
on a terrestrial species might include temperature, rainfall and other 
traditional measurements from the region being studied, but it might 
also include ocean temperature, and tidal and current flow data that 
may directly or indirectly influence the region, and it may also 
include weather patterns and pollution counts, all of which may be 
derived from observatories geographically far away that are owned and 
operated by other organizations. The ability to interact with and 
integrate data from multiple observatories that cross scientific, 
geographical, and administrative domains is an increasing requirement 
for environmental scientists today and presents a number of additional 
challenges with respect to coordination, standardization, and long term 
support for deployed cyberinfrastructure.
    Environmental observatories share many of the same general needs 
with other science domains including data storage and management, 
application codes, workflow systems to coordinate their research 
activities, and collaboration tools. However, it is the challenge of 
supporting potentially thousands of highly variable in situ sensors 
along with the need to manage and share them across vast geographical 
distances and administrative boundaries that makes environmental 
observatories unique.
    The proposed Genome 10K project \22\ is an example of the future of 
genomic research. The authors of this project, which includes 
scientists from across the world, are proposing to dramatically 
increase the number of vertebrate genomes available to the research 
community. This is made possible by a dramatic drop in sequencing costs 
coupled with a corresponding increase in computing capability. The 
Genome 10K Community of Scientists propose to assemble and sequence a 
collection of some 16,203 representative vertebrate species spanning 
evolutionary diversity across living mammals, birds, non-avian 
reptiles, amphibians, and fishes. This will allow scientists, for the 
first time ever, to carry out a comprehensive studies of vertebrate 
evolution. Just as computers enabled the assembly and annotation of the 
human genome, supercomputers will be required to manage and analyze 
massive quantities of genomic data to achieve the goals of the Genome 
10K project.
---------------------------------------------------------------------------
    \22\ ``Genome 10K: A Proposal to Obtain Whole-Genome Sequences for 
10,000 Vertebrate Species,'' Journal of Heredity, November 6, 2009.
---------------------------------------------------------------------------
    Status of Cyberinfrastructure for Data-driven Discovery. The 
development of cyberinfrastructure for data-driven discovery is in its 
infancy. Within NSF, most of the activity in this area is being driven 
by large Major Research Equipment & Facilities Construction (MREFC) 
projects. Each of these projects is developing the cyberinfrastructure 
needed to accomplish its mission, relying to some extent on the 
cyberinfrastructure developed in other projects but often redeveloping 
cyberinfrastructure capabilities in slightly different guises. Since 
one of the major issues associated with cyberinfrastructure is the 
ongoing support and maintenance costs associated with the software, 
sharing cyberinfrastructure software, wherever feasible, will help keep 
these costs under control.
    More recently, NSF has created major programs that are focused 
largely on the development of the cyberinfrastructure needed to support 
data-driven discovery. These include the iPlant Collaborative,\23\ a 
project aimed at developing cyberinfrastructure to address a number of 
grand challenges in plant biology (Genotype to Phentotype in Complex 
Environments, Tree of Life for Plant Sciences, etc.), and DataNet (NSF 
07-601), which consists of several projects designed to explore 
different approaches to organizing, managing and preserving the data 
being created in scientific and engineering research.
---------------------------------------------------------------------------
    \23\ See http://www.iplantcollaborative.org/.
---------------------------------------------------------------------------
    One of the major cyberinfrastructure requirements for data-driven 
discovery is the availability of the required data storage capacity, 
computing resources and associated software. Although these needs could 
often be met by augmenting the resources available at the NSF-funded 
supercomputing centers, most major data-driven discovery 
projects, which usually have lifetimes measured in decades, are 
reluctant to use the centers because of their uncertain future (current 
Track 2 grants are only for four years and funding for the Track 1 
system expires in 2016). This is a lost opportunity for leveraging the 
expertise at and cost efficiency of the supercomputing centers.

Networking

    To first order, the cyberinfrastructure most needed by universities 
to participate in or benefit from NSF's high performance computing and 
data-driven discovery projects is adequate network bandwidth linking 
them to the relevant project sites. The nation's major research 
universities are partners in Internet2, which provides a national high 
performance network. In addition, the National LambdaRail, which is 
also owned by the U.S. research and education community, provides a 
testbed for research in the development and use of communication 
technologies. However, this does not mean that all universities and 
colleges have access to network bandwidth adequate for their 
participation in or interaction with the big computing and data 
projects, an imbalance that will become more acute as the data volumes 
increase.
    As comfortable as the situation may be now,\24\ at least for the 
nation's major research universities, the volume of data that will be 
generated over the next few years in high performance computing and 
data-driven discovery will far outstrip the capacities of the current 
networks. For example, many simulations on Blue Waters will generate 
multiple terabyte data sets with the total amount of data generated in 
a given project being measured in petabytes. Although NCSA can provide 
connectivity to Chicago at 100-400 gigabits per second (Gbps), the 
national networks passing through Chicago (or any other U.S. city for 
that matter) do not have the capacity to deliver these data streams to 
the researchers' home institutions. Separate from the capacity issue, 
the underlying communication architecture, services and networking 
technologies required by data intensive science are very different from 
those that support common consumer services. Common carriers have shown 
little interest in meeting the specialized requirements of scientific 
research communities.
---------------------------------------------------------------------------
    \24\ Some scientists note that the current ``favorable'' situation 
is deceptive. Because of bandwidth limitations, they note that many 
scientists are simply avoiding research practices that would stress the 
current networks.
---------------------------------------------------------------------------
    In this regard it is worthwhile to note the DOE-SC's ESnet is a 
welcome exception. ESnet connects more than 40 sites across the nation, 
including all of DOE-SC's major experimental and computing facilities. 
DOE-SC's new Science Data Network, which is a part of ESnet, provides 
services that are specifically targeted for data-intensive science. The 
SDN circuits provide a wealth of services that are invaluable to 
scientists who need reliable, high performance, end-to-end connections 
between two or more sites. ESnet received funding under the American 
Recovery and Renewal Act to develop and deploy a 100 Gbps network 
linking its open supercomputing centers in California, Illinois and 
Tennessee. This is the first step toward DOE-SC's vision of a 1 terabit 
per second (Tops) network linking its major facilities.
    Although communications bandwidth. is critical to participating in 
high performance computing and data-driven discovery, the TeraGrid's 
Campus Champions program \25\ has shown that access to local expertise 
is also critical. This program supports individuals on university 
campuses who are knowledgeable about the TeraGrid and who can help 
faculty and students apply for and make use of the resources and 
services available through the TeraGrid. Such programs are likely to be 
just as important for data-driven discovery as for high performance 
computing.
---------------------------------------------------------------------------
    \25\ See: https://www.teragrid.org/web/eot/
campus-champions.
---------------------------------------------------------------------------
    Status of Networking. NSF was one of the pioneers in establishing a 
national networking infrastructure, e.g., NSFnet and Mosaic (the first 
web browser, which was created at NCSA). However, its networking 
infrastructure support programs were eliminated several years ago. So, 
the nation's scientists and engineers must rely on commercial 
providers, research and education network providers such as Internet 2 
and National LambdaRail, and state governments for their communications 
needs. To date, these entities have been able to provide the bandwidth 
and connectivity needed by researchers.24 However, with the major new 
data-intensive research resources coming on line, this will no longer 
be adequate.
    Another major problem is that, to date, there has been little focus 
on improving end-to-end networking capabilities, i.e., providing high 
performance connections between the researcher's desktop or local 
compute/data cluster and large computing and data sites. Even if it 
appears that there is adequate network bandwidth between these two end 
points, a bottleneck, often, but not always, on the researcher's campus 
dramatically limits the network performance. We need to have a better 
understanding of the issues affecting end-to-end performance to enable 
researchers to interact with their ongoing research activities at the 
major facilities.
    There are steps that NSF could take to ensure that researchers in 
U.S. universities have the networking capacity and policies needed to 
support their research. NSF could begin by developing a high 
performance network connecting all of their major research facilities, 
observatories, and supercomputing centers, interconnecting this network 
with those serving other major federal research facilities, e.g., 
ESnet, as needed by the academic research community. There are many 
advantages that will accrue from connecting NSF's large experimental 
and observational facilities with its computing and data facilities, 
especially if the future of these centers were secure. In addition, NSF 
could undertake pilot projects to obtain a better understanding of the 
problems limiting high performance end-to-end connections between 
researchers/small research groups and its major research facilities. 
This would require close collaboration between groups providing 
national networking resources and campuses providing the ``last mile'' 
connection.

Education

    I would be remiss if I did not include a section on education in 
responding to the Subcommittee's request for information on the state 
of cyberinfrastructure at U.S. universities. Although not a part of the 
cyberinfrastructure per se, our ability to advance science and 
engineering using the national cyberinfrastructure requires a new 
generation of scientists and engineers who can contribute to and 
understand the use of the basic technologies involved in 
cyberinfrastructure and computational science and engineering and who 
can collaborate with colleagues in other fields to take full advantage 
of the extraordinary capabilities provided by this infrastructure. We 
need to define the core competencies important for the next generation 
of scientists and engineers, followed by the development of 
implementation plan(s) to affect the needed curriculum and course 
changes.
    The curriculum and course changes required to educate the next 
generation of research leaders is not obvious. Many schools have 
established graduate programs in computational science and engineering 
that supplement study in a discipline with courses in ,computer science 
and engineering and applied mathematics; see, e.g., the Graduate 
Program in Computational Science and Engineering at the University of 
Illinois at Urbana-Champaign.\26\ Such programs are invaluable in 
preparing students for future careers in computing- and data-intensive 
fields. But are they sufficient? And what about undergraduate 
education? At the rate that analog science is becoming digital science, 
what do we need to teach all undergraduates in science and engineering 
about computing and related technologies to prepare them for life and 
work in the 21st century. Through its investments in research and 
education, NSF can serve as a catalyst for this transformation.
---------------------------------------------------------------------------
    \26\ See: http://www.cse.illinois.edu/.
---------------------------------------------------------------------------
    In the Blue Waters Project, we are pursuing this goal through the 
Virtual School of Computational Science and Engineering,\27\ headed by 
Professor Sharon Glotzer at the University of Michigan. The Virtual 
School brings together faculty across the universities in the Great 
Lakes Consortium for Petascale Computation to address the unique 
opportunities and challenges associated with petascale computing and 
petascale computing-enabled science and engineering. The Virtual School 
supports the creation and integration of courses and curricula that are 
tailored to the educational needs of 21st Century scientists and 
engineers, delivered using 21st century instructional technologies. 
Although the Virtual School is initially targeting graduate-level 
education, efforts in undergraduate education will follow.
---------------------------------------------------------------------------
    \27\ See: http://www.greatlakesconsortium.org/education/
VirtualSchool/.

                   Biography for Thom H. Dunning, Jr.

ta

    Chairman Lipinski. Thank you, Dr. Dunning.
    I would like to thank all the witnesses for their 
testimony. It is good to see especially with the rescheduled 
hearing that we have a pretty good turnout here. This is 
something that I think is very important to discuss since it is 
really key to figuring out what is best for our 
competitiveness, where to invest, although there are always 
tradeoffs that need to be made in talking about bringing back a 
program that was ended. I think it does require a lot of 
thought, so we will move on to questioning right now, and as is 
my tradition, I will leave myself to last on my side, so I will 
begin by recognizing Dr. Baird for five minutes, and I want to 
ask members to try to keep it down to the five minutes to at 
least get us through the first round of questions. I will be 
tight in trying to keep it to that. Dr. Baird
    Mr. Baird. Thank you, Mr. Chairman. I want to begin by 
echoing the accolades for our good friend and colleague, Dr. 
Ehlers. He served this committee and this country 
extraordinarily well, and it has been a privilege to serve with 
him. As I have said when people speak about my retirement, I am 
not dead yet, so I am sure, Vernon, the rest of us have a lot 
of good work to do but, Vern, thank you for your work.
    Thanks to the witnesses for their great comments and for 
your service to your schools and communities. I am very 
troubled by this issue of the infrastructure deficit that you 
talked about. Dr. Tolbert, you mentioned it; Dr. Raymond and 
Mr. Horvath, and then Dr. Dunning talked about sort of a 
different kind. There is one kind of deficit which seems to be 
the maintenance backlog, what are we doing to keep our existing 
facilities up to date, and then Dr. Dunning talked about the 
ability to keep up with what is happening in the other 
direction. Let me start with this maintenance issue. Dr. 
Tolbert, I think you said a $100 million deficit in some 
fashion. Can you elaborate on that a little bit?
    Dr. Tolbert. I think you may be referring to the $200 
million building renewal----
    Mr. Baird. I have cut your deficit in half. That is the 
kind of effective legislation I----
    Dr. Tolbert. Thank you very much. The $200 million number 
is actually a lower limit. It is actually much more than that. 
The State of Arizona owns the buildings, but it is up to the 
universities to keep them in shape and actually to try to keep 
them upgraded so that they are appropriate for the research in 
the fields they were built for. As state funds coming to the 
university have declined and private funding is scarce, this 
has become a real issue. In particular, donors are interested 
in funding new buildings when they have the funds. Right now 
isn't a good time for that. But when they have the funds, they 
are interested in new buildings. It is very hard to keep our 
older buildings in good working order. It is a little bit like 
the Nation's highway system. We built a lot of great 
universities some decades ago and now many of our science 
buildings are deteriorating.
    Mr. Baird. That is why I asked. I actually serve on the 
Highway Transportation Committee as well, and we talk a lot 
about the fiscal deficit in this country. It is about $1.3 
trillion per year. There is an existing infrastructure deficit 
exceed $1 trillion in highway, roads, et cetera, and now we are 
hearing about academic buildings.
    Dr. Raymond, you talked about a pretty large number in your 
case as well. What is your situation?
    Dr. Raymond. Our situation is very similar to the 
University of Arizona. The $100 million number that I gave you 
would be required to bring our buildings up to the minimum 
standards. If we brought them up to good standards, it would be 
closer to $200 million. The State of South Carolina provides a 
lump-sum appropriation to our university, which primarily is 
used to pay the light bills and salaries of support staff. 
There is very little left at the end of the day to take care of 
these older buildings, and we rely heavily on clinical revenue 
from our hospital, which obviously is diminishing these days, 
and from philanthropy to try to keep those buildings in good 
shape.
    Mr. Baird. Mr. Horvath, do you want to comment from your 
institution's perspective?
    Mr. Horvath. Sure. It is similar to the other two 
institutional examples. We have an overall facility deficit 
across the university of about $1 billion, and if you look at 
our annual capital plan, about 50 percent of the funding that 
we are investing in facilities and infrastructure is really 
being targeted to try to reduce that deferred maintenance 
backlog, and roughly 40 to 50 percent of that total relates to 
research, buildings and facilities which tend to be more 
intensive from a cost perspective.
    Mr. Baird. These are three of our major universities in the 
country and we are hearing now of deficits well exceeding $1 
billion from just three universities. Nationwide, this has got 
to be a serious problem, and you know, at a time when people 
are saying, well, we have got to cut taxes, et cetera, et 
cetera, there are going to be consequences, presumably for the 
quality of education received by your students and the 
competitiveness of their academic and intellectual products, I 
am guessing.
    Dr. Dunning, you are looking at it from the other side. One 
of the things this committee and others are worried about is 
the lag in U.S. competitiveness in other areas. If we don't 
make the investments, what happens to our supercomputer 
competitiveness?
    Dr. Dunning. Well, it is clearly going to decline if we 
don't make these types of investments. We see a resurgence of 
interest in high-performance computing and supercomputing 
across the world. The European Union has outlined a very 
aggressive program for building its supercomputing capability. 
The Japanese have long been in the game. But now we are also 
seeing significant advances in China and in other countries. So 
this is clearly an area in which competitiveness--although we 
are in a good position at the current time, competitiveness is 
something that will be challenging to maintain.
    Mr. Baird. Thank you, Mr. Chairman. Thanks to the 
witnesses.
    Chairman Lipinski. Thank you, Dr. Baird. That is a very 
good way to start us out here, and I will recognize Dr. Ehlers 
for five minutes.
    Mr. Ehlers. Thank you, Mr. Chairman, and I would also like 
to point out that Dr. Baird is not running for reelection 
either, so the message is clear.
    I appreciate your kind words for me, Dr. Dunning. I hope 
you feel that because we are both leaving, there is such a 
great need here for scientists that you will run for the 
Congress this fall and perhaps all of you carry that message 
back to your home institutions.
    Mr. Baird. Talk to me before you embark on that.
    Mr. Ehlers. But we desperately need more scientific and 
engineering talent in the Congress. We have most of it right 
here.
    Dr. Dunning, I just wanted to point out another dimension. 
A computer scientist in my district who teaches at the 
institution I used to teach at has come to see me and is very 
concerned about the declining enrollments in computer science 
nationwide, and in fact introduced a resolution to declare a 
week to emphasize computer science throughout the Nation and so 
forth. The professor who talked to me had very good statistics 
about what has been happening in declining enrollment but 
particularly what is happening in high schools; that they are 
now getting away from the subject, not getting students excited 
about computer science and all the issues related to it, and 
that doesn't necessarily even mean just programming or 
development, it means everything. And I am just wondering if 
you have encountered the same situation in your facility.
    Dr. Dunning. We have certainly seen that same situation. In 
fact, when we talk to our industrial partners--NCSA has a very 
large industrial partners program--one of their concerns is 
about the ability to really be able to recruit the type of 
workforce that they are going to need in the future where high-
performance computing has become a way of designing, for 
example, the next generation of products, products that reduce 
the environmental impact or can be done in a timely fashion. So 
we certainly see it at the university. We do see that it is 
slowing, which I think is good. One of the things that I think 
actually we find is an attractant to students interested in 
computer science is the applications of computer science. Many 
of them come in thinking of only a small number of applications 
of computer science, but basically what they aren't realizing 
is, all of science is open to a computer scientist, ranging all 
the way from astrophysics to zoology if they want to understand 
how they take the skills they learn as a computer scientist and 
really participate in this new era of digital science.
    Mr. Ehlers. Thank you for those comments. That reinforces 
what I have been told and what I have been trying to change 
here, and I hope you can get the rest of the computer science 
world excited about this.
    Dr. Dunning. So do we.
    Mr. Ehlers. That may be one of my retirement projects.
    Dr. Dunning. Good.
    Mr. Ehlers. Thank you. I yield back.
    Chairman Lipinski. Thank you, Dr. Ehlers.
    I will now recognize Ms. Fudge.
    Ms. Fudge. Thank you, Mr. Chairman, and thank all of you 
for being here.
    I represent northeast Ohio. We have one of the largest 
research institutions, Case Western Reserve University, and 
over the years Case has found time and found a way to work more 
collaboratively with our large research hospitals, the 
Cleveland Clinic and university hospitals in particular, and so 
just understanding how we function in our area. My question to 
any of the panel members is that certainly we all recognize 
that states are struggling and the current economy is really 
not very good, but the fact is that many states began divesting 
in higher education long before the economic downturn, and 
industry as a whole invests very little in higher education and 
university research, even though both the states and industry 
rely very heavily on strong universities for their own growth 
and success. I just want to know from you or hear from you what 
you are doing as a university community to lobby your states, 
your local industry and some public support to sustain your 
investments in higher education that will benefit all of us. 
Anyone? Dr. Tolbert, thank you.
    Dr. Tolbert. Thank you very much, Congresswoman Fudge, for 
your question. This is a really important role that I think we 
all play. Certainly in my role at the University of Arizona as 
a senior administrator, I must be making that kind of argument 
all the time, both to the state and locally, actually, to local 
government as well, and also to the private sector. We are 
working hard to broaden and deepen our relationships with the 
largest corporations that we work with. We have a huge Raytheon 
presence in Tucson. Raytheon Missile Systems is based in Tucson 
and the University of Arizona provides more Raytheon employees 
than any other university in the country. That is really 
important to us, and we are trying to find ways that we can 
increase the benefit to both by increasing the number of 
internships, for example, available to our students and so on.
    But the argument that has to be made, I think, is, 
importantly, not one about short-term gain only. It is that 
this is a long process we are engaged in. We need long-term 
relationships with the private sector. We need the state to 
understand our long-term needs and not to say that we can 
enhance economic development tomorrow, but that if we want 
Arizona to pull itself out of this deep, deep economic slump, 
we must have research universities doing research but also 
educating students in a research-oriented environment.
    Ms. Fudge. Thank you.
    Dr. Raymond, in your testimony you mentioned that there are 
certain initiatives that the State of South Carolina has done 
to expand state research capabilities. Can any of the other 
witnesses talk about how their universities are linking 
research investments to regional economic goals? And I am 
taking this line because I went to Ohio State University, which 
is a very well known research institution as well, but we have 
them all over the country, and certainly if their reliance is 
going to be on us, we can't do all of them. So what we are 
doing as a community is to try to find other sources, but as 
well as to collaborate more, because I think that is where we 
are all going to have to be heading. Anyone? Dr. Raymond, if 
you would like to expand, or someone else on the panel?
    Dr. Raymond. Thank you. By the way, I was born in Akron and 
went to Ohio State.
    Ms. Fudge. All right. I knew there was something I liked 
about you.
    Dr. Raymond. South Carolina is trying to build critical 
mass through a collaboration between the four largest hospital 
systems in the state and the three research-intensive 
universities, Clemson, USC and MUSC, in which we all put at 
least $2 million a year into joint initiatives and recruitment 
of endowed professors to the state. This interdigitates well 
with the State of South Carolina's investment through the 
Centers of Economic Excellence program, that provides matching 
dollars for recruitment of new talent to the State of South 
Carolina. So we have really worked very well together. This 
initiative now is in its fifth year and it has spawned patient 
safety initiatives, public-private partnerships and 
partnerships with large pharmaceutical companies and device 
makers that wouldn't otherwise be interested in dealing with 
one of our universities, but the power of having a $10 billion 
budget and 40,000 employees can convince them that we are good 
partners.
    Ms. Fudge. Thank you very much. Yes, Dr. Horvath.
    Mr. Horvath. One of the things that I would point out just 
recently that we were successful with was the achievement of a 
grant through the Recovery Act that is going to enable us to be 
part of a 19-institution consortium in Pennsylvania to really 
expand and add robustness to our computing networks throughout 
the State of Pennsylvania. It is a large-scale project. There 
is private money that will be involved in that and it is 
really, I think, one of the indications of a new way in which 
we are going to have to collaborate across institutions to make 
some of the needs or address some of the needs that we have in 
our states and locally.
    Ms. Fudge. Thank you, Mr. Chairman. I yield back.
    Chairman Lipinski. Thank you, Ms. Fudge.
    I will recognize Ms. Johnson.
    Ms. Johnson. Thank you very much, Mr. Chairman, and let me 
express my appreciation to Dr. Ehlers. I hate to see Dr. Ehlers 
retire. He has been a friend of mine on this committee now for 
as long as he has been here, and I have relied heavily on your 
knowledge. What I would like to ask of the distinguished 
witnesses--and thank you for being here--are you seeing any 
more student readiness as a research university? That is number 
one. And then, we have some major research departments, and I 
don't know how well they coordinate--NIH, the National Science 
Foundation, Department of Defense, Energy, Ag, Education. Is 
there enough collaboration between them to maximize research 
dollars?
    Dr. Tolbert. I will answer your first question, and then I 
would also like to say something about your second question, 
Congresswoman. The first question had to do with readiness. We 
do not see improving readiness. In fact, we are having to put 
in place or deciding to put in place a lot of sort of one-on-
one assistance to students to try to help students who really 
aren't quite ready to find their way into the university but 
also we are greatly growing our relationship with the community 
colleges in the state so that we can do two-plus-two programs 
that help with readiness.
    And the other question?
    Ms. Johnson. The other question is, we have major research 
departments here and some of it could overlap or some could be 
in coordination with medical schools and others, and I wonder 
if there is enough collaboration between these departments to 
maximize what dollars we do have for research.
    Dr. Tolbert. I think that is an extremely good question 
because the NIH has done something interesting. They have 
several programs working across institutes within the NIH. I am 
a neuroscientist. My funding has almost all been through the 
NIH. And that is I think working very well for NIH. Is there 
enough coordination across agencies? I am sure we would all 
argue no. It is very difficult but more coordination would be 
better. And one of the things that I suggested was a committee 
be appointed that would look across federal agencies at this 
issue of funding of research infrastructure for just that 
reason.
    Ms. Johnson. Thank you.
    Anyone else?
    Dr. Dunning. Let me take a slightly different tack on your 
question, and that is readiness of graduate students to 
participate in research that is heavily involved in computing, 
and I would say again, we are not properly educating the next 
generation of scientists and engineers to really understand the 
major role that computing and information science is going to 
play in their careers, especially when you consider that their 
careers are going to last 30 to 40 years into the future.
    Ms. Johnson. Thank you very much, Mr. Chairman.
    Chairman Lipinski. Thank you, Ms. Johnson.
    I will now recognize myself for five minutes. I think we 
had very good questions so far. I wanted to follow up a little 
bit on some of the questions and answers. All of you talked 
about what really is the deficit in terms of the infrastructure 
right now, and we know that a lot of the university 
laboratories and other research facilities were built 30 or 
more years ago, even before some of the requirements of modern 
science and engineering researchers were even thought of. Two 
questions that I want to ask related to that. The first is for 
all of you. Do you find it difficult to compete for faculty and 
graduate students because of the infrastructure or lack of 
infrastructure that you have? Whoever wants to offer. Dr. 
Tolbert.
    Dr. Tolbert. Thank you very much, Mr. Chairman. That is 
such an interesting question. What we find is that in the areas 
in which we have built recent research buildings, or research 
laboratories when in older buildings, we are much more 
competitive. Of course, we have put the dollars into areas 
where we wanted to grow, but for instance, we have a life 
sciences building that is doing a wonderful job for us of 
attracting new faculty and also very good undergraduate and 
graduate students. So as we put facilities in place, I think I 
can argue pretty strongly that we find that we are more 
competitive, not only in bringing people but in keeping them, 
which is also a very important issue.
    Chairman Lipinski. Anyone else want to take a stab at that? 
Dr. Raymond.
    Dr. Raymond. Well, I would echo our experience. When we 
have put up new buildings for our health professions, the 
quality of students and faculty that we are able to attract and 
retain goes up enormously. So if you superimpose that on our 
background, I would say that we probably are having difficulty 
recruiting top talent to some of the buildings and facilities 
that maybe aren't as up to date.
    Chairman Lipinski. I had a dual listening session last week 
in northern California with a number of universities, and one 
thing that they had mentioned and I had raised earlier was 
losing out to other countries. Have you seen, especially 
graduate students who come from other countries, a greater 
likelihood that they return to their home country than, say, 
was the case five, ten years ago? Dr. Raymond?
    Dr. Raymond. I recently completed a trip to Shanghai and 
Guangzhou, China. We have a partner university there, Zhejiana 
University, and we have post-docs and faculty members who have 
been at our institution for 5 to 20 years who are now returning 
to China because they find that the facilities are as good or 
better than we have here and there is a lot of investment by 
the local government and the federal government in startup 
companies and in providing them with opportunities for 
entrepreneurship. So we have a very tangible loss in our own 
institution to China.
    Chairman Lipinski. Anyone else? Dr. Dunning?
    Dr. Dunning. Yes, another comment related to that is that 
the Chinese government is in the process of establishing three 
to five major supercomputing centers which are going to be 
partnerships between the provinces in China. And the federal 
government in China, and this clearly is attracting recent 
graduates and more senior professors back to China to 
participate in this effort.
    Chairman Lipinski. One thing that, certainly, as we look to 
the future and competitiveness, one thing that people often say 
to me is, well, we still have the greatest universities in the 
world in this country, and it certainly is a concern if we see 
that starting to go away and lose our competitiveness there.
    One other thing I wanted to--the second part I wanted to 
put out there is, are we losing productivity in our research 
because of the facilities? We talked about, and I mentioned at 
the beginning, there is a tradeoff. We would all love to fund 
this infrastructure but we know that there is not unlimited 
dollars, and I realize that, and I know bringing up the 
possibility of reauthorizing ARI would raise this issue but I 
wanted to do it. Is there a problem and do you see a potential 
problem, do you see it happening already of losing how 
productive the research is? We just keep putting the funding 
into research, and if we don't have the facilities, are we 
losing out on the value of that money? Mr. Horvath?
    Mr. Horvath. In older facilities that have more substantial 
needs, essentially what happens is, band-aids are applied 
rather than, perhaps, more fundamental changes or upgrades 
being made to building infrastructure or the specific 
instrumentation or equipment in a particular lab. I think if we 
had the flexibility and ability to make more fundamental 
investments in those types of facilities, I think we would see 
greater productivity and better cost efficiency as a result of 
being able to do those things.
    Chairman Lipinski. Dr. Raymond, did you----
    Dr. Raymond. Yes. Thank you, Mr. Chairman. We just did a 
study on a research facility, a very simple metric, dollars per 
square foot of federal grants earned by our faculty, and it is 
no coincidence that the newer and more modern buildings are 
populated by faculty that are more productive. And furthermore, 
to operate the buildings costs a lot more for the older 
facilities so our investment is not getting the same kind of 
returns in the older facilities that it is in the newer 
facilities, both in terms of faculty productivity and in terms 
of the cost to keep those buildings operating.
    Chairman Lipinski. Dr. Tolbert.
    Dr. Tolbert. We found exactly the same thing. I would just 
like to echo what Dr. Raymond said, same statistics.
    Chairman Lipinski. Thank you. With that, I will yield back 
my time. Dr. Ehlers, do you have further questions? I will 
yield five minutes to Dr. Ehlers.
    Mr. Ehlers. Thank you, Mr. Chairman. I just want to clear 
up a few questions that I have.
    Mr. Horvath, in your testimony you had a list of federal 
regulatory and legislative changes that you say directly affect 
the conduct and management of research under federal grants and 
contracts, and some of them are self-evident but you mentioned 
the Lobbying Disclosure Act and a section of the Higher 
Education Act relating to the reporting of foreign gifts, 
contracts and relationships. I am curious, how do those impede 
your efforts?
    Mr. Horvath. I think they were all essentially listed to 
provide some background to how the regulatory environment has 
changed in which universities have to operate, and those 
changes have occurred since the imposition of the indirect cost 
cap, the 26 percent administrative cap back in 1991. All of 
those things have come together to create greater complexity 
and encouraged us to add processes and staff to be able to 
respond to those things. So I think the point of that was to 
say that the regulatory environment has become much more 
complex in an era when we are constrained in terms of the 
amount of costs that we can recover to respond to those.
    Mr. Ehlers. Now, is it because the researchers have to take 
time from their research to meet their requirements or because 
your office is burdened with having to respond?
    Mr. Horvath. It is basically the administrative 
responsibilities that faculty, for example, now have to 
shoulder as a result of some of those new requirements, are not 
able to be recovered and take some of their time away from 
conducting research in their laboratories.
    Mr. Ehlers. Thank you.
    Then Dr. Raymond, you got the NSF research infrastructure 
improvement grant. That sounds very exciting. What is the value 
of the grant, and I am wondering how the money will be 
distributed. You mentioned it expands on the contribution of 
the non-research but I would like to see you expand on the 
contribution of the non-research incentive intensive 
institutions, the smaller universities, community colleges, 
minority institutions. I have often wondered how we could 
develop an effective method of doing that nationwide. Could you 
give me a little background on that, please?
    Mr. Raymond. I would love to have the opportunity to update 
you in three or four years when we can actually measure 
outcomes, but for the first time we have the senior research 
universities willing to sit at the same table with community 
colleges, technical colleges and four-year institutions to 
develop an overall plan to address the deficiencies in the 
pipeline for STEM disciplines in our universities, and they are 
also now reaching out to K-12 in the state, so I hope that this 
provided the seed corn, so to speak, to really have a good 
outcome.
    Mr. Ehlers. Thank you. I yield back, Mr. Chairman.
    Chairman Lipinski. Thank you, Dr. Ehlers. I was just 
discussing the timing of the grant announcements from the 
Recovery and Reinvestment Act, and I know that is going to be 
relatively soon. Have your universities applied for grants? Dr. 
Tolbert?
    Dr. Tolbert. Thank you, Mr. Chairman. We have never seen so 
much grant-writing activity as we have seen in the last year. 
In fact, it has been a huge burden on our sponsored projects 
office, which reports to me, another one of these hidden costs. 
People have been working nights and weekends to handle all the 
grant proposal activity that has happened. We have won $83 
million in Recovery Act funds, and the reporting now of the 
activities also is another hidden cost, administrative cost. We 
are delighted to have those funds. Most of the funds are for 
research projects. About $6 million of that is for 
infrastructure, buildings and equipment, but most of it is for 
the direct costs of research. Our faculty are delighted, but 
the research reporting requirement on my office is now huge. 
That is the downside of being very successful. We are delighted 
with the Recovery Act.
    Chairman Lipinski. In terms of the research infrastructure 
grants, we are supposed to hear soon on a lot of those. Some of 
those will have to wait until September. Mr. Horvath?
    Mr. Horvath. We have submitted applications for grants on 
two specific facilities. One is the renovation of one of our 
health and human development buildings, and the second for 
biological research lab and building, and we are hoping to hear 
positive news sometime in the next few days or few weeks.
    Chairman Lipinski. Dr. Tolbert?
    Dr. Tolbert. And we have learned that we will be getting an 
NIH grant for $15 million for a research support building in 
the new arm of our college of medicine in Phoenix.
    Chairman Lipinski. Dr. Raymond?
    Dr. Raymond. We have also recently learned that we received 
an $8 million grant to substantially renovate an older facility 
on campus to bring the microbiology and immunology labs up to 
speed. One of the problems we had in the building, it is humid 
in Charleston and we were having difficulty maintaining 
positive air pressure so there was mold growing in some of the 
micro and immuno labs so you can't do bacteriology research 
with fungus growing in there, so this will help us to bring our 
facilities up to snuff.
    Chairman Lipinski. Thank you. I just wanted to conclude, I 
want to ask Dr. Dunning a couple things. How much does the NCSA 
spend helping its user community?
    Dr. Dunning. I would say on the average probably around $4 
to $5 million a year. That is primarily staff time. I must say 
not all of that is funded by the National Science Foundation, 
however; a good portion of that is funded by the State of 
Illinois and the University of Illinois.
    Chairman Lipinski. Do you think there should be more 
funding to help users?
    Dr. Dunning. We are seeing large growth in the number of 
communities that need high-performance computing to be able to 
move into the new areas they want to go into and to solve the 
types of problems that they are encountering. I think the only 
way to successfully move those communities in that direction is 
through strong user support.
    Chairman Lipinski. I have heard that there are issues in 
terms of being able to--people who can really use those 
facilities being able to understand how they can use them, the 
ability to--not even having the knowledge of what is possible 
to do and how to do it. Do you find that?
    Dr. Dunning. Some of the communities are coming from a very 
low base. One of the environmental communities we worked with a 
couple of years ago was using Excel spreadsheets as the means 
of storing and analyzing all of their data. That works for a 
while, but when you start accumulating the quantities of data 
that these environmental groups are now talking about being 
able to gather by the sensor arrays and other devices that they 
have, that won't work in the future, and so there is a 
tremendous education and training aspect that goes along with 
the user support. One of the things that we find really 
critical for our user support is to have a staff member that 
actually has some training in either that area or related 
areas, so he understands the scientific objectives and 
techniques that community is using, because the more you know 
about the community, the better you're able to move them down 
the pipeline toward being able to use these much larger 
computing systems.
    Chairman Lipinski. Thank you. I will yield back. Dr. 
Ehlers, any more questions, closing comments?
    Mr. Ehlers. I have taken too much time already.
    Chairman Lipinski. I want to thank all of our witnesses for 
testifying before the Subcommittee today. The record will 
remain open for two weeks for additional statements from the 
Members and for answers to any follow-up questions the 
Committee may ask of the witnesses.
    Again, I thank the witnesses all for their testimony. 
Certainly as we move forward with NSF reauthorization, with 
America COMPETES reauthorization, I welcome any more comments 
that you have. Your input is critical if we are going to do 
this right, and we are really going to make sure we are putting 
our resources in the best possible place for our universities, 
for research and for our competitiveness.
    With that, the witnesses are excused and the hearing is now 
adjourned.
    [Whereupon, at 3:27 p.m., the Subcommittee was adjourned.]

                                   
