[Congressional Record Volume 150, Number 28 (Monday, March 8, 2004)]
[Senate]
[Pages S2308-S2309]
From the Congressional Record Online through the Government Publishing Office [www.gpo.gov]

      By Mr. BINGAMAN (for himself and Mr. Alexander):
  S. 2176. A bill to require the Secretary of Energy to carry out a 
program of research and development to advance high-end computing; to 
the Committee on Energy and Natural Resources.
  Mr. BINGAMAN. Mr. President, I am pleased to introduce, along with 
Senator Alexander, the High-End Computing Revitalization Act of 2004. 
High-end computing, also known as high performance computing or 
supercomputing, is a critical component to the scientific advances, 
defense capabilities, and commercial competitiveness of the United 
States in the 21st century. Several recent developments in high-end 
computing have stimulated a re-examination of current U.S. policies and 
approaches. These developments include: 1. the deployment of Japan's 
Earth System Simulator, which now occupies the number one position on 
the Top 500 list of the world's fastest computers; 2. concerns about 
the difficulty in achieving substantial fractions of peak hardware 
computational performance on high-end systems; and 3. the ongoing 
complexity of developing, debugging, and optimizing applications for 
high-end systems. In addition, there is growing recognition that a new 
set of scientific and engineering discoveries could be catalyzed by 
access to very-large-scale computer systems--those in the 100-teraflop 
to petaflop range. Lastly, the National Academies of Sciences and 
Engineering, the Office of Science and Technology Policy's High End 
Computing Revitalization Task Force, and the national security 
community have each released interim or final comprehensive reports 
expressing serious concern over the current U.S. position in high-end 
computing research. Without government support, market forces are 
unlikely to drive sufficient innovation in high-end computing, because 
the private sector would not capture the full value of its innovations 
on a short enough time scale.
  In supercomputing, innovation is important in architecture, in 
software, and in application strategies and solution methods. The 
coupling of these aspects is equally important. Major architecture 
challenges stem from the uneven performance scaling of different 
components. In particular, as the gap between processor speeds, memory 
bandwidth, and memory and network latency increases, new ideas are 
needed to increase bandwith and mitigate latency. Additionally, as new 
mechanisms are introduced to address those issues, there is a need for 
ways to supply a stable software interface that facilitates exploiting 
hardware performance improvements while the changes in mechanism. A new 
large-scale computer system exceeding the capability of Japan's Earth 
Simulator would provide an excellent test-bed for promising new 
advancements in these areas as well as provide science, industry, and 
defense with an extraordinarily powerful new tool for advancing the 
interests of the United States.
  The need for software innovation is motivated by its role as an 
intermediary between the application, the problem being addressed, and 
the architectural platform. Innovation is needed in the ways that 
system software manages the use of hardware resources, such as network 
communication. New approaches are needed for ways in which the 
applications programmer can express parallelism at a level high enough 
to reflect the application solution and without platform-specific 
details. Novel tools are needed to help application-level software 
designers development their solutions at a more abstract and problem-
specific level. Software technology is also needed to lessen future 
dependence on legacy codes. Enough must be invested in the creation of 
advanced tool and environment support for new language approaches so 
that users can more readily adopt new software technology.
  Importantly, advances in algorithms can sometimes improve performance 
much more than architectural and other software advances do. A center 
for high-end computing software would aid immensely in spurring 
innovation in this underdeveloped research area, an aid in coordinating 
the federal government's efforts with industry, academia, and between 
its national laboratories.
  The department of Energy (DOE) and its Office of Science research 
programs are uniquely qualified to lead research in these fields. They 
have played an important role in the development of high-end computing, 
networking, and information technology. These capabilities have been 
readily accessible to the U.S. scientific community for a diverse set 
of grand challenge scientific computational problems. Contributions by 
the DOE include pioneering the concept of remote, interactive access to 
supercomputers (developing the first interactive operating system for 
supercomputers, establishing the first national supercomputer center); 
developing the mathematical foundations for high performance computing 
with numerical linear algebra libraries used worldwide; leading the 
transition to massively parallel supercomputing by developing software 
to allow processors to communicate with each other; and contributing to 
the development of the Internet with software that dramatically speeds 
up the transmission of messages.
  Many challenges are associated with modeling complex physical, 
chemical, and biological phenomena, especially on massively parallel 
computers with peak speeds in hundreds of teraflops. These challenges 
include the management and analysis of petabyte-scale data sets. A 
program to address these challenges will require multi-disciplinary 
collaborations between theoretical and computational scientists, 
computer scientists, and applied mathematicians at universities, 
national laboratories, and industry. Such a program will enhance the 
ability of the DOE to meet its mission goals and advance the state of 
the art for the U.S. economic and industrial base in the fields of 
energy, genetics, pharmaceuticals, nanotechnology, chemical processing, 
electronics, geology, and transportation. This bill will be a major 
step toward addressing today's greatest needs and, to creating the 
high-wage jobs of the future.
  I ask unanimous consent that the text of the bill be printed in the 
Record.
  There being no objection, the bill was ordered to be printed in the 
Record, as follows:

                                S. 2176

       Be it enacted by the Senate and House of Representatives of 
     the United States of America in Congress assembled,

     SECTION 1. SHORT TITLE.

       This Act may be cited as the ``High-End Computing 
     Revitalization Act of 2004''.

     SEC. 2. FINDINGS.

       Congress finds that--
       (1) high-end computing is a critical component of the 
     scientific advances, defense capabilities, and commercial 
     competitiveness of the United States in the 21st century;
       (2) with the deployment of the Earth System Simulator in 
     Japan, the United States no longer has a clear lead in high-
     end computing worldwide;
       (3)(A) promising new architectures should be developed that 
     increase memory and network bandwidth, minimize latency, and 
     coordinate the architectures' various components to maximize 
     application performance; and
       (B) it is recognized that different architectures may be 
     better suited to different applications;
       (4)(A) software that improves efficiency on and 
     accessibility to high-end systems should be developed; and
       (B) this development effort should include research in 
     optimal algorithms, programming environments, tools, 
     languages, and operating systems for high-end computing, in 
     collaboration with architecture development efforts;
       (5) without government support, market forces are unlikely 
     to drive sufficient innovation in high-end computing, because 
     the private sector would not capture the full value of its 
     innovations on a short enough time frame; and
       (6) having played an important role in the development of 
     high-end computing, networking, and information technology, 
     the Department of Energy, and the research programs of the 
     Office of Science of the Department, are particularly 
     qualified to lead research in those fields.

[[Page S2309]]

     SEC. 3. DEFINITIONS.

       In this Act:
       (1) High-end computing system.--
       (A) In general.--The term ``high-end computing system'' 
     means a computing system with performance that substantially 
     exceeds commonly available systems.
       (B) Inclusions.--The term ``high-end computing system'' 
     includes a system described in subparagraph (A) that is based 
     on a variety of architectures, including vector, 
     reconfigurable logic, streaming, processor-in-memory, and 
     multithreading architectures.
       (2) Institution of higher education.--The term 
     ``institution of higher education'' has the meaning given the 
     term in section 101(a) of the Higher Education Act of 1965 
     (20 U.S.C. 1001(a)).
       (3) Secretary.--The term ``Secretary'' means the Secretary 
     of Energy, acting through the Director of the Office of 
     Science of the Department of Energy.
       (4) Ultrascale scientific computing capability.--The term 
     ``ultrascale scientific computing capability'' means a 
     computing capability supporting open scientific research in 
     the United States that is at least 100 times such computing 
     capability in existence on the date of enactment of this Act.

     SEC. 4. HIGH-END COMPUTING SYSTEMS PROGRAM.

       (a) In General.--In addition to any other authority 
     provided by law, the Secretary shall carry out a program of 
     research and development (involving software and hardware) to 
     advance high-end computing systems.
       (b) Duties.--In carrying out the program, the Secretary 
     shall--
       (1) support both individual investigators and 
     multidisciplinary teams of investigators;
       (2) conduct research in multiple architectures, including 
     vector, reconfigurable logic, streaming, processor-in-memory, 
     and multithreading architectures;
       (3) conduct research in software development on optimal 
     algorithms, programming environments, tools, languages, and 
     operating systems for high-end computing systems, in 
     collaboration with architecture development efforts;
       (4) in accordance with subsection (c), develop, plan, 
     construct, acquire, or operate equipment or facilities for 
     the use of investigators conducting research and development 
     on an ultrascale scientific computing capability;
       (5) support technology transfer to the private sector and 
     others in accordance with applicable law; and
       (6) ensure that the program is coordinated with relevant 
     activities in industry and other Federal agencies, including 
     the National Nuclear Security Administration, the National 
     Science Foundation, the Defense Advanced Research Projects 
     Agency, and the National Security Agency.
       (c) Ultrascale Scientific Computing Capability.--
       (1) In general.--As part of the program carried out under 
     this Act, the Secretary shall develop, plan, construct, 
     acquire, or operate a coordinated set of facilities for 
     investigators to develop an ultrascale scientific computing 
     capability for--
       (A) scientific research and development using high-end 
     computing systems; and
       (B) developing potential advancements in high-end computing 
     system architecture and software.
       (2) Administration.--In carrying out this subsection, the 
     Secretary shall--
       (A) support multiple high-end computing system 
     architectures; and
       (B) conduct research on the basis of proposals (including 
     proposals that are submitted by industry, institutions of 
     higher education, national laboratories, or any Federal 
     agency) for research on problems that would particularly 
     benefit from large computing power, even as the reliability 
     of new hardware and software components are being evaluated.
       (d) High-End Software Development Center.--
       (1) In general.--As part of the program carried out under 
     this Act, the Secretary shall develop, plan, construct, 
     acquire, or operate at least 1 High-End Software Development 
     Center.
       (2) Duties.--A Center shall concentrate efforts to develop, 
     test, maintain, and support optimal algorithms, programming 
     environments, tools, languages, and operating systems for 
     high-end computing systems.
       (3) Staff.--A Center shall include--
       (A) a regular research staff, to create a centralized 
     knowledge-base for high-end software development; and
       (B) a rotating staff of researchers from other institutions 
     and industry to assist in the coordination of research 
     efforts and promote technology transfer to the private 
     sector.
       (4) Use of expertise.--The Secretary shall use the 
     expertise of a Center to assess research and development in 
     high-end computing system architecture.
       (5) Location.--The location of a Center shall be determined 
     by a competitive proposal process administered by the 
     Secretary.
       (e) Peer Review.--Each grant, contract, cooperative 
     agreement, and financial assistance awarded under this 
     section shall be made only after independent peer review.
       (f) Classified Research or Facilities.--No funds under this 
     section may be used to directly support classified research 
     or facilities.

     SEC. 5. AUTHORIZATION OF APPROPRIATIONS.

       (a) In General.--In addition to amounts made available for 
     high-end computing systems under other provisions of law, 
     there are authorized to be appropriated to the Secretary to 
     carry out this Act--
       (1) $150,000,000 for fiscal year 2005;
       (2) $155,000,000 for fiscal year 2006;
       (3) $160,000,000 for fiscal year 2007;
       (4) $165,000,000 for fiscal year 2008; and
       (5) $170,000,000 for fiscal year 2009.
       (b) Ultrascale Scientific Computing Capability.--Of the 
     funds made available under subsection (a), $100,000,000 is 
     authorized to be appropriated for each fiscal year to carry 
     out section 4(c).
       (c) High-End Software Development Center.--Of the funds 
     made available under subsection (a), $10,000,000 is 
     authorized to be appropriated for each fiscal year to carry 
     out section 4(d).

                          ____________________