Defense IRM: Investments at Risk for DOD Computer Centers (Letter Report,
04/04/97, GAO/AIMD-97-39).
Pursuant to a congressional request, GAO reviewed: (1) the Department of
Defense's (DOD) plans to consolidate, outsource, and modernize its
computer center operations; and (2) whether DOD has an effective
framework in place for making and executing these decisions. GAO did not
validate the accuracy of the information provided by DOD on the numbers
and costs of computer centers, the alternative analyses, funding plans,
and processing capacities.
GAO noted that: (1) DOD has recognized the need to continue reductions
in the cost of its computer centers' operations through consolidation,
modernization, and outsourcing, but it has not yet established an
effective framework for making these decisions; (2) this framework would
include departmentwide policies and procedures critical to the success
of its efforts to improve computer centers; (3) these policies and
procedures would establish targets for how many computer centers DOD
actually needs, define how mainframes and mid-tier computer operations
should be consolidated, and identify the numbers and skill mix of staff
that are required to operate the centers, and what constitutes an
optimum computer center; (4) DOD also has no mechanism for ensuring that
the best money-saving opportunities have been considered by the
individual services and components or that consolidation efforts will
conform to federal requirements or even meet the needs of DOD as a
whole; (5) as a result, DOD services and components have developed
individual strategies for consolidating and modernizing their computer
centers that are inconsistent and contradictory to DOD as a whole and
may well cause DOD to waste millions of dollars in computer center
expenditures; (6) in addition, the consolidation strategies of the
military services and DOD components did not always fully address
critical planning elements required by the Office of Management and
Budget (OMB) requirements that could help reduce the risk of waste,
including alternative analyses, high-level implementation plans, and
funding plans; (7) further, GAO found that the OMB and departmental
guidance, particularly addressing mid-tier computer centers, was
unclear; (8) this resulted in inconsistent interpretation and reporting
for these centers; (9) therefore, OMB and DOD do not have assurance that
the computer consolidation strategies are sound; (10) without better
management over the implementation of its computer center strategies,
DOD at best will only achieve optimization at the component level and
forgo optimization for DOD as a whole; and (11) moreover, DOD's chief
information officer (CIO) is now required by the Paperwork Reduction Act
of 1995, Clinger-Cohen Act of 1996, and the Fiscal Year 1997 DOD
Authorization Act to develop and implement a plan for a management
framework with policies and procedures as well as effective oversight m*
--------------------------- Indexing Terms -----------------------------
REPORTNUM: AIMD-97-39
TITLE: Defense IRM: Investments at Risk for DOD Computer Centers
DATE: 04/04/97
SUBJECT: Privatization
ADP
Strategic information systems planning
Interagency relations
Federal agency reorganization
Information resources management
Computer services contracts
Defense economic analysis
Management information systems
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO report. Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved. Major **
** divisions and subdivisions of the text, such as Chapters, **
** Sections, and Appendixes, are identified by double and **
** single lines. The numbers on the right end of these lines **
** indicate the position of each of the subsections in the **
** document outline. These numbers do NOT correspond with the **
** page numbers of the printed product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
** A printed copy of this report may be obtained from the GAO **
** Document Distribution Center. For further details, please **
** send an e-mail message to: **
** **
** **
** **
** with the message 'info' in the body. **
******************************************************************
Cover
================================================================ COVER
Report to the Chairman, Committee on National Security, House of
Representatives
April 1997
DEFENSE IRM - INVESTMENTS AT RISK
FOR DOD COMPUTER CENTERS
GAO/AIMD-97-39
Defense IRM
(511348)
Abbreviations
=============================================================== ABBREV
BRAC - Base Realignment and Closure
C3I - Command, Control, Communications, and Intelligence
CIO - Chief Information Officer
DeCA - Defense Commissary Agency
DIA - Defense Intelligence Agency
DIS - Defense Investigative Service
DISA - Defense Information Systems Agency
DLA - Defense Logistics Agency
DOD - Department of Defense
DSWA - Defense Special Weapons Agency
MIPS - millions of instructions per second
NIMA - National Imagery and Mapping Agency
OMB - Office of Management and Budget
Letter
=============================================================== LETTER
B-271572
April 4, 1997
The Honorable Floyd D. Spence
Chairman, Committee on National Security
House of Representatives
Dear Mr. Chairman:
This report responds to your request that we review the Department of
Defense's (DOD) plans to consolidate, outsource, and modernize its
computer center operations and assess whether DOD has an effective
framework in place for making and executing these decisions. DOD and
the individual military services have consolidated a number of their
computer centers in recent years and contracted with the private
sector for information processing services that were previously
performed in-house. These actions were part of an effort to find
better and less costly ways of meeting Defense information processing
needs. DOD has recognized that there are opportunities for further
consolidations since, according to DOD reports, about 40 percent of
its computer centers still fall well below governmentwide minimum
processing capacity targets.
You also asked us to perform a subsequent detailed review of the
plans of the Defense Information Systems Agency (DISA) to consolidate
and modernize DOD's megacenters, or central processing facilities.
As agreed with your office, we will be reporting separately on this
issue.
RESULTS IN BRIEF
------------------------------------------------------------ Letter :1
DOD has recognized the need to continue reductions in the cost of its
computer centers' operations through consolidation, modernization,
and outsourcing, but it has not yet established an effective
framework for making these decisions. This framework would include
departmentwide policies and procedures critical to the success of its
efforts to improve computer centers. These policies and procedures
would establish targets for how many computer centers the Department
actually needs, define how mainframe and mid-tier computer operations
should be consolidated, and identify the numbers and skill mix of
staff that are required to operate the centers, and what constitutes
an optimum computer center. Defense also has no mechanism for
ensuring that the best money-saving opportunities have been
considered by the individual services and components or that
consolidation efforts will conform to federal requirements or even
meet the needs of the Department as a whole.
As a result, Defense services and components have developed
individual strategies for consolidating and modernizing their
computer centers that are inconsistent and contradictory to the
Department of Defense as a whole and may well cause Defense to waste
millions of dollars in computer center expenditures. For example,
the consolidation plans of DOD's primary information processing
service provider, the Defense Information Systems Agency, assume
services and components will send DISA additional information
processing business. Most of the services and components, however,
are not planning to do so. Thus, DISA's planned investment, and in
turn DOD's, in providing new information processing services may be
wasted. In addition, the consolidation strategies of the military
services and DOD components did not always fully address critical
planning elements required by the Office of Management and Budget
(OMB) requirements that could help reduce the risk of waste,
including alternative analyses, high-level implementation plans, and
funding plans. Further, we found that the OMB and departmental
guidance, particularly in addressing mid-tier computer centers, was
unclear. This resulted in inconsistent interpretation and reporting
for these centers. Therefore, OMB and DOD do not have assurance that
the computer consolidation strategies are sound.
Without better management over the implementation of its computer
center strategies, Defense at best will only achieve optimization at
the component level and forgo optimization for the Department as a
whole. Moreover, Defense's chief information officer is now required
by the Paperwork Reduction Act of 1995, the Clinger-Cohen Act of
1996, and the Fiscal Year 1997 Department of Defense Authorization
Act to develop and implement a plan for a management framework with
policies and procedures as well as effective oversight mechanisms for
ensuring that major technology related efforts, such as the computer
center consolidations, conform to departmentwide goals.
SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :2
In conducting our review, we reviewed and analyzed various DOD
computer center consolidation plans and reported costs and assessed
how these plans met OMB's Bulletin 96-02 requirements. We also
compared DOD plans and practices to the practices and strategies
employed by private-sector companies we visited during our review
that have successfully consolidated computer centers. In addition,
we met with consultants who advise computer center managers on
improving services and with the General Services Administration's
Office of Governmentwide Policy and Federal Systems Management
Center. We conducted numerous interviews with DOD officials to
discuss their approach to consolidating and modernizing computer
centers. We also discussed with OMB officials the OMB Bulletin
governing computer centers and their views on DOD responses. Details
of our scope and methodology are included in appendix I.
We did not validate the accuracy of the information provided by DOD
on the numbers and costs of computer centers, the alternatives
analyses, funding plans, and processing capacities. Our work was
performed from March 1996 through January 1997 in accordance with
generally accepted government auditing standards.
The Department of Defense provided written comments on a draft of
this report. These comments are presented and evaluated at the end
of this letter and are reprinted along with our more detailed
evaluation in appendix II. The Office of Management and Budget
provided oral comments on a draft of this report which are
incorporated in the report as appropriate and discussed at the end of
this letter.
BACKGROUND
------------------------------------------------------------ Letter :3
The federal government owns hundreds of computer centers that perform
such services as processing agency software programs, providing
office automation and records management, and assisting in the
management of wide area computer networks. In recent years, the
federal government has recognized that most of these centers operate
below optimum capacity, use outdated technology, and perform
redundant services. It has concluded that it can achieve significant
dollar savings and operational efficiencies by consolidating computer
centers or by acquiring its information processing services from the
private sector.
In 1993, the Vice President's National Performance Review\1
recommended that the federal government take advantage of evolving
technology and begin consolidating and modernizing its computer
centers to reduce the duplication in information processing services
and decrease information processing costs. To help implement this
recommendation, a committee formed by the Council of Federal Data
Center Directors\2 recommended that the Office of Management and
Budget establish operating capacity targets for the consolidated
centers and that federal agencies follow an approach successfully
used by private-sector companies and other government agencies to
plan, implement, and optimize their own computer centers.
The Committee's recommendations formed the basis of OMB guidance to
promote computer center improvements and consolidations, which was
issued in October 1995. This guidance, OMB Bulletin 96-02,
Consolidation of Agency Data Centers, called on agencies to (1)
reduce the number of their computer centers, (2) collocate small and
mid-tier computer platforms in larger computer centers, (3) modernize
their remaining centers in order to improve the delivery of services,
and (4) outsource information processing services to other federal or
commercial computer centers when aggregate computer center capacities
were below minimum target sizes. Table 1 lists OMB's specific
requirements.
Table 1
OMB Requirements for Computer Centers
Deadline Agency action
------------------ --------------------------------------------------
March 1, 1996 Submit an inventory of agency computer centers,
including
--each computer center's name, location, and
mission;
--the basic hardware configuration of the centers,
including the type of mainframe processors used
and the use of small and mid-tier processors
(those machines that fall in the range between a
work station and mainframe and provide such
services as client servers and network
controllers);
--the numbers and skill mix of staff; and
--costs for hardware, software, staffing,
utilities, communications, and contract services.
June 3, 1996 Submit a computer center consolidation strategy
for either (1) meeting minimum computer center
sizes established by OMB or (2) describing how the
agency planned to outsource its processing to
other federal or commercial computer centers. This
strategy was to include
--an alternatives analysis reflecting the
technical feasibility and cost-effectiveness of
alternatives--including outsourcing;
--an architecture design, or technical solution,
identifying the receiving and closing data centers
and workload realignment as well as the
communications architecture;
--a high-level implementation approach identifying
major consolidation tasks and presenting a
schedule, milestones, and resources;
--a funding plan identifying and forecasting costs
associated with the consolidation process and
funding requirements for all major tasks
associated with the consolidation; and
--exceptions that could not be included in the
consolidation plan.
September 2, 1996 Submit a detailed implementation plan for
consolidating or outsourcing, including a detailed
technical architecture, a transition plan, a
security and disaster recovery plan, a human
resources plan, and an acquisition plan.
June 1998 Complete consolidations.
----------------------------------------------------------------------
OMB allowed agencies considerable discretion as to which data centers
they chose to retain and close so long as their consolidation
scenario was cost-effective and minimal data center target sizes were
met. The target sizes OMB set were based on a standard industry
measure for information processing: millions of instructions per
second, or MIPS. OMB asked that centers using IBM mainframe
computers operate at 325 MIPS and centers using UNISYS operating
systems\3 operate at 225 MIPS.\4 Further, OMB permitted agencies to
justify not consolidating centers that fell below the target size if
a particular center
-- had a staff of less than five full-time employees,
-- housed scientific processors and would otherwise be at least 90
percent of the minimal target size, or
-- housed a large number of small and mid-tier processors and would
otherwise be at least 90 percent of the minimal target size.
OMB's guidance is in keeping with recent congressional initiatives
that focus on strengthening the planning and management of
information technology efforts. In implementing the Paperwork
Reduction Act and the Clinger-Cohen Act, OMB requires that
information technology investments support core/priority mission
functions and that they be undertaken by the requesting agency
because no alternative private-sector or governmental source can
efficiently support the function. These laws and OMB guidance also
require agencies to establish an enterprisewide investment approach
to information technology that includes selecting, controlling, and
evaluating investments as part of an integrated set of management
practices designed to link investments to organizational goals and
objectives. Further, the National Defense Authorization Act for
Fiscal Year 1997 requires the Secretary of Defense to report the
Department's plan for establishing an integrated framework for
management of information resources within the Department by March 1,
1997.\5
OMB's guidance is also in keeping with the approach private-sector
companies have taken in successfully consolidating and modernizing
their own computer processing centers. We analyzed successful
consolidation and modernization efforts carried out by three
corporations and learned that they believed it was necessary to
implement their strategies from a corporatewide perspective, rather
than have separate components of their companies consolidate and
modernize their own centers.\6
These companies also ensured that from the outset of their
consolidation efforts, they had clear and consistent policies and
procedures governing how computer center services would be improved.
This guidance spelled out such things as what constitutes an optimum
computing center in terms of capacity and staff, what skills were
needed to operate the centers, what cost and performance goals were
relevant for the centers, and which services could be outsourced. In
setting capacity goals, the private-sector companies we visited also
generally attempt to reach targets that are substantially higher than
the ones set by OMB--from 1,000 to 3,500 MIPS.
In addition, we learned that private-sector companies we visited
during this review established strong oversight processes for
ensuring that their computer center decisions were based on accurate,
complete, and current information on cost, schedules, benefits, and
risks; that all valid options for their computer center services were
fully addressed; and that their current services were correctly
benchmarked against comparable services.
--------------------
\1 Report of the National Performance Review: Creating a Government
That Works Better and Costs Less, September 7, 1993.
\2 The Council is a nonprofit organization that promotes the
administration of information technology and computer centers.
\3 Operating systems are the software that controls the execution of
programs. An operating system may provide such services as resource
allocation, scheduling, input/output control, and data management.
\4 OMB set its MIPS targets below the private sector's generally
accepted targets because it believed that these targets were more
achievable for government agencies and would result in significant
savings.
\5 This plan was submitted to the Congress on March 14, 1997.
\6 We identified the corporations with successful computer center
efforts through discussions with private-sector consultants and
Defense officials.
DOD RECOGNIZES BENEFITS OF
FURTHER CONSOLIDATING,
MODERNIZING, AND OUTSOURCING
COMPUTER CENTERS
------------------------------------------------------------ Letter :4
In 1996, Defense reported to OMB that it owned 155 computer centers
that perform a variety of information processing related services for
the services and components. Among other things, the centers run
software programs developed by the military services and various
Defense components and provide information security services,
customer help desk services, and records management services.
Sixteen of these centers are central processing facilities known as
megacenters and are owned by DISA. The remaining 139 centers are
service- or component-unique centers.
DOD also reported that it was continuing to further optimize and
standardize its computer centers operations as part of departmentwide
and intra-agency consolidations that had started in 1990 and continue
today. Defense has recognized that these computer centers have been
operating inefficiently and that they need to adopt new technologies
and address the increasing loss of in-house technical expertise in
order to continue supporting the Department's large and complex
information infrastructure. We agree with DOD that there are still
many opportunities for savings. In fact, table 2 shows that many of
these reported centers operate below the minimum processing capacity
targets established by OMB for government-owned computer centers and
thus are good candidates for consolidation or outsourcing.
Table 2
Status of DOD Computer Centers at the
Time of OMB Submissions (as of September
30, 1995)
(Dollars in thousands)
Number meeting
Service Number of private-sector
or centers meeting minimum capacity
componen Computer center Number of OMB criteria for standard of 1000
t costs\a computer centers consolidation MIPS
-------- ---------------- ---------------- ---------------- ----------------
DISA $531,121 28 16 0
Other 60,287\b 26 16 0
compone
nts
Army 135,321 59 3\c 0
Navy 56,724\d 14\e 14\e 0
Air 132,035 28 13\f 0
Force
================================================================================
Total $915,488 155 62 0
--------------------------------------------------------------------------------
\a Costs as defined in OMB Bulletin 96-02.
\b Does not include cost recovery for Defense Intelligence Agency.
These costs are classified.
\c Army reported that 56 of its 59 centers, including its small and
mid-tier centers, are used for remote or local processing or for
unique service missions (e.g., National Guard, command and control)
and therefore should not be considered for consolidation.
\d The Navy reported the costs of all its computer centers.
\e The Navy did not report the number of all its computer centers,
only the mainframe centers were reported in its aggregate center
inventory.
\f According to Air Force officials, 11 of its 13 small and mid-tier
data centers are exempt from consolidation because they meet the
provisions of the national security exemption within the
Clinger-Cohen Act, which they believe exempts combat ammunition
systems from consolidation.
Source: DOD's OMB Bulletin 96-02 inventory submissions. We did not
independently verify this information.
As table 2 indicates, 62 of DOD's 155 computer centers--about 40
percent--met OMB criteria for possible consolidation based on
processing capacity targets. However, we believe that these numbers
could be higher. As indicated in the table notes, many small or
mid-tier centers were not considered as candidates for consolidation.
According to OMB officials responsible for implementing the Bulletin,
these centers should have been included unless otherwise exempted.
Further, private-sector and government-sector studies have found that
larger facilities allow organizations to economize on floor space,
staff, and operating expenditures, and smaller centers tend to be
cost-inefficient.\7
Before OMB issued its computer center Bulletin 96-02, DOD had
determined, based on industry practices, that consolidation would
''position DOD to more effectively support common data processing
requirements across services by leveraging information technology and
resource investments to meet multiple needs.'' Since 1990, the
Department has initiated and completed multiple intra-agency
consolidations. In 1993, the megacenters were established as a
result of (1) DOD's base closures and (2) other consolidation and
cost reduction efforts. In establishing these centers, DOD expected
to change its information processing environment from one that was
stovepiped, or confined to individual military services and
components, to one that supported information sharing DOD-wide.
Accordingly, since 1990, DOD consolidated its computer center
operations by moving the workload and equipment from 194 DOD computer
centers into 16 DISA megacenters by fiscal year 1996, reporting a
reduction in processing costs of over $500 million.
After these consolidations, DOD initiated several studies that looked
into the question of whether the remaining megacenters should be
further consolidated, modernized, or outsourced. One study--done by
the Defense Science Board on the question of outsourcing DOD
functions in general--reported in August 1996 that processing at
computer centers was more expensive than at private-sector computer
centers and it recommended that DOD computer center services be
outsourced.\8
A second study--done by a private contractor on the question of
outsourcing, modernizing, and consolidating DISA's megacenters for
the Assistant Secretary of Defense for Command, Control,
Communications, and Intelligence (C3I) in 1996--concluded that
further consolidation and outsourcing of megacenter operations was
feasible.\9 The contractor reported that the megacenters' life cycle
(10 years) cost could be cut by more than a billion dollars if the
megacenters were consolidated to 6 from 16 and if certain computer
center services--such as the customer help desk and those services
associated with day-to-day operation of the centers--were fully
outsourced.\10
The Undersecretary of Defense, Comptroller, was also directed to
submit a report on the feasibility of outsourcing DOD's megacenters
to the House Appropriations Committee\11 by January 1, 1996. In this
report, which was submitted to the Congress on December 26, 1996, the
Comptroller largely agreed with the recommendations made by the
contractor study described above and supported DISA's proposed
management plan to implement those recommendations. Some of our
concerns with this plan, which is DISA's consolidation strategy, are
discussed in more detail in the next section of this report. In
addition, as noted in the beginning of this report, we will be
reporting separately on our detailed review of DISA's plans.
--------------------
\7 Council of Federal Data Center Directors' Federal Data Center
Consolidation Committee, Consolidation of Federal Data Centers
(February 1995) and KPMG Peat Marwick, Best Data Center Practices
(December 1994).
\8 Report of the Defense Science Board Task Force on Outsourcing and
Privatization, dated August 1996.
\9 Strategy Options for Defense Information Services, Final Report,
Coopers and Lybrand Consulting, dated February 1996.
\10 We did not validate the savings estimate in this study.
\11 As directed by the House and Conference Reports accompanying the
Fiscal Year 1996 Defense Appropriation Act.
CONSOLIDATION STRATEGIES ARE
INCONSISTENT AND FALL SHORT OF
MEETING OMB REQUIREMENTS
------------------------------------------------------------ Letter :5
While DOD and its components have made progress in consolidating and
finding opportunities to optimize and outsource many of the functions
of its computer center operations, DOD is still missing opportunities
to achieve even greater savings under its current approach. The
Defense leadership has chosen to allow the individual military
services and components to carry out their computer center
consolidation and modernization efforts independent of any
departmentwide framework. In fact, the Assistant Secretary of
Defense for C3I, as part of his guidance when forwarding the OMB
Bulletin, stated "that each Service and Defense Agency has the
flexibility to reduce its data centers in a manner that is consistent
with the DOD Component's goals, management philosophy, and
environment as long as such reductions occur within the framework of
the OMB guidelines." This decision has resulted in inconsistent and
contradictory strategies which fall short of meeting OMB's
requirements and what we believe to be the intent of OMB's Bulletin.
We also learned that some of the inconsistency and incompleteness of
reported plans and strategies was caused, in part, by DOD's broad and
inconsistent interpretations of the OMB Bulletin. Appendix IV
provides a detailed analysis of how the military services and
components responded to OMB's Bulletin. The following discussion
highlights our findings.
STRATEGIES VARY WIDELY
---------------------------------------------------------- Letter :5.1
As shown in the two tables that follow, the computer center
consolidation plans of the individual military services and
components submitted to OMB to date vary widely. For example, the
Air Force, the Army, three Navy commands, and three Defense
components plan to further consolidate in-house, while other parts of
the Army and Navy, as well as the Defense Investigative Service, are
choosing to keep their computer center operations in-house without
further consolidation. Many of the strategies reflect a move toward
mid-tier solutions without considering the potential for
consolidation. Only the National Imagery and Mapping Agency and
parts of the Navy chose to outsource (to DISA) their center
operations. Further, just two services--the Air Force and the
Army--considered inter-service consolidation of their respective
computer centers within the Pentagon, and this action was already
underway prior to the OMB Bulletin. Table 3 describes the approaches
the services and components have decided on. Table 4 compares the
strategies.
Table 3
Computer Center Consolidation and
Modernization Approaches
Agency Approach
------------------ --------------------------------------------------
DISA Continue to consolidate to 16 megacenters and then
to further consolidate the 16 megacenters and
outsource certain services. Attract new mainframe
and mid-tier processing business from other
components and military services.
Air Force Continue its ongoing modernization, moving to a
client-server architecture and outsourcing efforts
at three of its computer centers, and consolidate
its Pentagon Center with the Army's. Identify
computer centers that can transfer applications
from mainframe to other environments. Consolidate
or outsource any remaining computer centers.
Army Retain local area networks and file servers at the
installation level. Consolidate its center in the
Pentagon with Air Force. Continue with in-house
operations for its Personnel Information Systems
Command and Army Reserve Personnel Data Centers,
pending further analysis on migration to client-
server architecture.
Navy Transfer the workload from eight centers to DISA.
Retain the Navy Medical Information Management
Center computer in-house pending further review of
alternatives. Transfer the remaining processing
that is below MIPS target levels to new, in-
house, mid-tier computers.
Defense Commissary Moving to a client-server architecture.
Agency
Defense Continuing to consolidate and modernize in-house
Intelligence systems and outsource some services.
Agency
Defense Its center exceeds minimum consolidation
Investigative thresholds. Moving to a client-server
Service architecture.
Defense Logistics Mainframe processors consolidated. Continuing in-
Agency house consolidation of mid-tier centers. Explored
using other federal data centers for potential
outsourcing; analysis supported retaining the work
in-house.
National Imagery Will transfer new system work to DISA's
and Mapping Agency megacenters.
Defense Special In final stages of consolidating its three
Weapons Agency computer centers into two in-house computer
centers.
----------------------------------------------------------------------
Table 4
Comparison of the Consolidation and
Modernization Strategies
Maintain Outsource
Agenc status Consolidate Outsource selected Attract new
y quo in-house centers services business
----- ------------- ------------- ------------- ------------- -------------
DISA X X X
Air X X
Force
Army X X
Navy X X X
Defen X
se
Commi
ssary
Agenc
y
Defen X X
se
Intel
ligen
ce
Agenc
y
Defen X
se
Inves
tigat
ive
Servi
ce
Defen X
se
Logis
tics
Agenc
y
Natio X
nal
Image
ry
and
Mappi
ng
Agenc
y
Defen X
se
Speci
al
Weapo
ns
Agenc
y
--------------------------------------------------------------------------------
We found that some of these strategies had contradictions that might
well have been prevented had Defense better coordinated its computer
center efforts. For example, as table 3 notes, the Department's
primary information processing service provider, DISA, intends to
modernize and consolidate its megacenters and begin to offer mid-tier
processing services to attract additional business from the services
and components. DISA believes that significant reductions in the
cost of operations could be achieved and that much of DOD's computer
processing is well suited for consolidation to DISA's computer center
operations.
However, it is clear from the strategies described above that most of
the services and agencies are not considering sending additional
business to DISA, and DISA has no authority to require the services
and components to make such transfers. The Army, the Air Force, and
the Defense Logistics Agency, for example, do not plan to increase
their use of DISA services. Together, about $385,000 of the reported
$915,500 spent on computer center operations is outside of DISA.
CONSOLIDATION STRATEGIES DO
NOT MEET OMB REQUIREMENTS
---------------------------------------------------------- Letter :5.2
Most of the consolidation strategies submitted by the military
services and components to OMB failed to fully address all of the
planning elements addressed in the Bulletin. For example, some
services and components did not provide sufficient information to
show that they had (1) performed thorough analyses of their planned
options, (2) demonstrated that they had ensured that they have the
correct technical solutions to their computer center operations, (3)
prepared even a high-level implementation approach to the major tasks
associated with consolidation, or (4) provided estimates on how much
it will cost to consolidate and modernize. Therefore, DOD and OMB do
not have assurance that the services and components are addressing
these critical planning elements in carrying out their strategies or
that the approaches they have chosen are sound.
Table 5 summarizes how the individual services and components
responded to the OMB Bulletin. ("N" meaning they didn't respond, "Y"
meaning they did respond, "P" meaning they partially responded, and
"N/A" meaning not applicable.) The table illustrates that the Defense
Intelligence Agency, Defense Investigative Service, and Defense
Special Weapons Agency were the only DOD components in compliance
with all of the OMB requirements. The Army was the only other
component to have submitted a complete alternatives analysis for its
computer centers. The table also shows that the Air Force, DISA, the
Defense Commissary Agency, and the Defense Logistics Agency either
partially addressed or did not address the requirements. As noted
earlier, a more detailed analysis of the responses is provided in
appendix IV.
Table 5
GAO Analysis of How Services and
Components Responded to OMB Requirements
for Consolidation Strategies
Air
Forc
OMB requirements DISA e Army Navy DeCA DIA DIS DLA NIMA DSWA
-------------------- ---- ---- ---- ---- ---- ---- ---- ---- ---- ----
Alternatives P P Y P N * Y P P Y
analysis reflecting
technical
feasibility and
cost-effectiveness
of alternatives
including
outsourcing?
Architecture design N P Y P N * * P Y Y
identifying
receiving and
closing centers and
workload realignment
as well as
communications
architecture?
High-level N P P P N * * P P Y
implementation
approach identifying
major consolidation
tasks and presenting
a schedule,
milestones, and
resources?
Funding plan N P P P N * * P N Y
identifying and
forecasting costs
associated with the
consolidation
process and funding
requirements for all
major tasks
associated with the
consolidation?
Exceptions N/A Y Y N/A N Y Y Y N/A Y
identified that
could not be
included in the
consolidation plan?
--------------------------------------------------------------------------------
*Agency reported that its computer centers met OMB's consolidation
threshold for processing capacity.
Legend
Y=Responded.
N=Did not respond or response was inadequate.
P=Partial response provided. (Partial responses are explained in
more detail in appendix IV.)
N/A=Did not request or did not apply.
As reflected in table 5, we determined that the DOD submissions did
not always comply with the OMB requirements. When we discussed this
with DOD officials, they said that their submissions did not always
describe their consolidation plans for their non-mainframe computer
centers because some of the components believed that the guidance
only applied to mainframes and others believed that the guidance did
not apply to actions already underway and approved through DOD's life
cycle management process. These officials had interpreted the
Bulletin as requiring that non-mainframe computer centers be included
in the inventory but not in the consolidation strategies, unless they
affected the center's meeting the minimum target size. When we
discussed this with OMB officials, they disagreed with DOD's
interpretation. They stated that the Department's non-mainframe
centers also should have been addressed in both the inventories and
the consolidation strategies.
We also asked OMB officials why they required submissions from each
of the military departments and one from DOD. OMB officials told us
that they required four separate submissions based on their
interpretation of the Paperwork Reduction Act of 1995, which defined
DOD as the Department of Defense and the three services. However,
they further stated they would have preferred to receive from DOD a
departmentwide inventory, consolidation strategy, and implementation
plan that clearly reflected a departmentwide analysis and direction
for DOD decisions on computer centers. We believe such a
departmentwide approach is consistent with the intent of the Bulletin
and the Clinger-Cohen Act to ensure that opportunities to consolidate
centers among services and components were maximized. Instead, these
officials stated that OMB received separate and conflicting responses
that failed to provide a clear view of consolidation across
components. OMB officials further stated they had difficulty
determining how many centers DOD currently had and planned to have
after the consolidations.
When we discussed the multiple submissions from DOD with Defense
officials, the Assistant Secretary of Defense for C3I acknowledged
that the military services and DOD components had developed
individual plans. However, he believed that separate plans were
allowed by the OMB Bulletin and that OMB did not request a
departmentwide strategy or plans. However, the Assistant Secretary
agreed that the Department needs a departmentwide policy guidance and
framework as DOD seeks additional opportunities for economies and
efficiencies in its data center operations. The Assistant Secretary
also agrees that future decisions should be based on sound business
analyses and that the Clinger-Cohen Act provides a context and
leverage for these guidelines.
DOD LACKS CRITICAL
DECISION-MAKING TOOLS FOR
CONSOLIDATION EFFORTS
------------------------------------------------------------ Letter :6
Although DOD has been consolidating its computer centers since 1990,
we found, and the Assistant Secretary of Defense for C3I agreed, that
DOD lacks several decision-making tools that are imperative to any
computer consolidation and modernization effort. First, it has not
set targets or established policy for basic things, such as how many
computer centers the Department actually needs, the numbers and skill
mix of staff that are required to operate the centers, and what
constitutes an optimum computer center. It also has no mechanism for
ensuring that the best money-saving opportunities have been
considered by the individual services and components or that
consolidation efforts will conform to federal requirements or even
the needs of the Department as a whole. As discussed earlier in this
report, private companies we visited during our review found that
setting such targets--through policies and procedures--and oversight
mechanisms were key to the success of their consolidation efforts.
Without them, DOD will have difficulty identifying problematic
strategies and preventing some of its computer center investments
from being wasteful.
DOD HAS NO DEPARTMENTWIDE
POLICIES AND PROCEDURES ON
COMPUTER CENTERS
---------------------------------------------------------- Letter :6.1
The private companies we visited during our review found it necessary
to direct their computer center consolidation efforts from a
corporatewide perspective and to clearly delineate the makeup and
number of centers that the companies were aiming for. More
specifically, these companies established policies that defined what
constituted an optimum computer center in terms of processing
capacity and the numbers and skill mix of its staff; how many centers
the corporation needed; what computer center functions were so
critical to carrying out the company's mission that they could not be
outsourced; what cost and performance goals were relevant for the
centers; and how the centers should be compared, or benchmarked, to
more successful operations. They also established corporatewide
procedures for implementing these policies.
During our review, we also learned that DOD visited private
companies, including the ones we visited. DOD officials benchmarked
this industry experience to determine how best to prepare, justify,
and implement prior departmentwide efforts to consolidate and
standardize computer centers from 1990 to 1994. For example, DOD
learned examples of private-sector criteria that could be used to
select megacenters and the level of processing capacity and
expandable floor space these centers should have. However, it did
not use the lessons learned from these visits to prepare
departmentwide policies and procedures.
As a result, individual services and components do not have a
consistent basis for determining what constitutes an optimum center;
what their performance or staffing targets should be; or which
functions are inherently governmental or can be outsourced. For
example, these services and components do not have departmentwide
targets that they can set as goals for the processing capacities of
their mainframe or mid-tier centers.
In a March 1996 report\12 on DOD's acquisition of computer centers,
DOD's inspector general specifically noted that Defense lacked
complete "policies and procedures on acquiring and managing the
proper mix of mainframe and mid-tier computers to process corporate
data" and that without such policies and procedures--especially those
for mid-tier processors--DOD's potential for acquiring excess
computer processing capabilities increases. The Inspector General
also noted that if DOD would coordinate its processing needs it
could, among other things, (1) take advantage of the open systems\13
infrastructure concept to resolve operational problems, (2) better
track and report information management costs on a DOD-wide basis,
(3) better manage the transition from existing outdated systems to
migration systems, and (4) improve management of computer security.
DOD agreed with the Office of the Inspector General that it should
establish procedures for evaluating and providing corporate
information processing and storage requirements on a DOD-wide basis
rather than on an individual program basis. However, DOD noted that
it should proceed with care in implementing this recommendation
because of its implications for centralized management and control.
According to officials in the office of the Assistant Secretary for
C3I, DOD plans to determine if, and to what extent, it has a mid-tier
computing problem before issuing policies and procedures to address
that problem.
Under the Paperwork Reduction Act of 1995 and the Clinger-Cohen Act,
passed in 1996, the Assistant Secretary of Defense for Command,
Control, Communications and Intelligence, as DOD's Chief Information
Officer (CIO), is supposed to develop and implement management policy
and procedures to ensure that major information technology related
efforts conform to departmentwide goals. In a memorandum dated
November 6, 1995, the Assistant Secretary expressed an intent to
monitor the consolidation initiatives to (1) ensure consistent
interpretation and implementation of OMB Bulletin 96-02 across the
Department, (2) ensure that consolidation efforts are consistent with
the DISA plans, and (3) identify issues and develop strategies for
resolving them quickly. Accordingly, the Assistant Secretary set up
an advisory group to provide policy guidance for the Department's
efforts to consolidate and outsource computer center operations.
However, this group has not yet prepared this critical guidance nor
has it been effective in achieving its stated monitoring objectives.
--------------------
\12 Acquisition of Computers That Process Corporate Information
(DOD/OIG Report 96-081, March 5, 1996).
\13 Computer applications that can communicate with each other across
a network and across computer applications that use a common
operating system interface.
DOD HAS NO MEANS OF ENSURING
THAT BEST OPPORTUNITIES ARE
IDENTIFIED
---------------------------------------------------------- Letter :6.2
Under the Clinger-Cohen Act, the Secretary of Defense, with the
advice and assistance of the CIO, is responsible for establishing a
mechanism for ensuring that the military services and components have
considered the best investment options and consolidation efforts that
will meet the needs of the Department as a whole. In its guidance to
agencies on evaluating information technology investments, OMB
suggests that such a mechanism take the form of an investment review
board, or senior management team, that would review information
technology funding decisions. In their decision-making process, the
team would consider such things as strategic improvements versus
maintenance of current operations, new projects versus ongoing
projects, risks, opportunity costs, and budget constraints. The
Assistant Secretary C3I also charged the advisory group discussed
above with the responsibility for providing oversight for computer
center consolidation efforts. Yet, to date, neither the advisory
group nor any other DOD component has provided this oversight. The
Assistant Secretary C3I believes that his authority as DOD's chief
information officer for providing such oversight has been
strengthened by the Clinger-Cohen Act. However, he also believes
that his office lacks the staff and departmentwide support to
establish such oversight.
Without this important oversight mechanism, DOD does not have a means
for assessing whether the individual services and components
considered the cost-effectiveness and technical feasibility of their
computer center alternatives from a departmentwide perspective and
whether their implementation approaches, schedules, and funding plans
are realistic. This also precludes Defense from having an
opportunity to review the consistency of the individual plans and
identify and recommend areas where even more monetary and efficiency
gains could be achieved through inter-service and component efforts.
CONCLUSIONS
------------------------------------------------------------ Letter :7
Without better coordination and oversight of computer center
consolidation efforts, the best Defense can hope to achieve from its
computer center consolidations is optimization at or below the
component level. It will certainly miss out on the chance to ensure
that the most investment worthy opportunities are identified and
implemented, such as those that involve services and components
merging their computer centers. Moreover, millions of dollars in
computer center investments and operating expenses may well end up
being wasted since individual components and services are planning
without departmentwide information processing needs in mind and
without the benefit of clearly defined organizationwide policies and
procedures for the consolidation efforts and effective oversight
mechanisms. Having centralized coordination for computer center
optimization efforts and strong policies, procedures, and oversight
were integral to the success of the corporations we visited in their
efforts to consolidate computer centers. They should be for Defense
as well.
RECOMMENDATIONS
------------------------------------------------------------ Letter :8
We recommend that the Secretary of Defense direct the Department's
Chief Information Officer to develop an integrated, departmentwide
plan for improving the cost and operations of its computer centers.
Until this plan is approved by the Secretary, we further recommend
that the Secretary of Defense limit any capital investments in the
Department computer centers to investments that meet critical
technology needs to operate the DOD computer centers. The
Department's CIO should certify that these investments comply with
departmentwide goals and technical standards.
We also recommend that as a basis for this plan and for future
decisions concerning consolidation, modernization, and outsourcing of
computer centers, Defense's Chief Information Officer develop
policies and related procedures that address the following:
(1) what constitutes an optimum computer center in terms of
processing capacity and staff numbers and skills;
(2) how many computer centers are needed;
(3) which of its computer center operations are inherently
governmental and/or require component-unique centers solutions and
thus cannot be consolidated or outsourced;
(4) how DOD should compare its computer center services with those of
other public-sector and private-sector services in terms of cost,
speed, productivity, and quality of outputs and outcomes; and
(5) which cost and performance goals are relevant for comparing
departmentwide alternatives.
We also recommend that Defense's Chief Information Officer establish
or incorporate within its existing processes, as practical, the
necessary oversight to ensure that the above recommended
departmentwide plan and future computer center consolidation,
modernization, and outsourcing decisions (1) are being developed in
accordance with the above policies and procedures, (2) are based on a
sound analysis of alternatives, and (3) consider the goals and needs
of the entire department.
Finally, we recommend that the Director of the Office of Management
and Budget (1) clarify its Bulletin, particularly in regard to
mid-tier consolidation criteria and its intent to have an integrated
Department of Defense submission and (2) require the Department of
Defense to replace its prior multiple submissions in response to this
new guidance with an integrated departmentwide submission that
contains a departmentwide inventory of computer centers, a
departmentwide consolidation strategy, and a departmentwide
implementation plan.
AGENCY COMMENTS AND OUR
EVALUATION
------------------------------------------------------------ Letter :9
The Department of Defense provided written comments on a draft of
this report. OMB provided us with oral comments. DOD concurred with
our recommedation on providing oversight over its computer center
efforts and partially concurred with our recommendation to develop
policies and procedures to guide computer center decisions. However,
DOD did not concur with our recommendation to limit any capital
investments in the Department's computer centers until an integrated,
departmentwide consolidation plan is prepared. Defense's response to
this report is summarized below, along with our evaluation. Appendix
II contains Defense's comments along with our more detailed
evaluation.
DOD agreed that it needs to develop a prudent framework for achieving
potential savings through its future computer center consolidation,
modernization, and outsourcing decisions. DOD added that it is
developing such a framework as part of its effort to implement the
Clinger-Cohen Act. DOD also questioned the need for an integrated
consolidation and outsourcing plan since the Department has already
consolidated many of its computer centers, with significant reported
savings, without such a plan. However, during our review, DOD
officials acknowledged that unlike prior consolidation efforts, DOD
has allowed the components considerable flexibility in their current
consolidation efforts, without strategic direction from the
Department. Thus, we continue to believe that an integrated,
departmentwide plan is needed to show that the Department's computer
center decisions reflect sound choices for meeting departmentwide
processing needs and not just those of the individual components.
In discussing our recommendation on developing policies and
procedures for making consolidation and outsourcing decisions, DOD
agreed that these are necessary. However, DOD believed that it
should complete its development of an integrated management framework
for implementing the Clinger-Cohen Act before developing the specific
policies and procedures we recommended. We are encouraged by the
Department's effort to begin to develop a management framework for
implementing the Clinger-Cohen Act, especially if it includes the
policies and procedures we recommend in this report. The report
detailing DOD's plans for this framework was submitted to the
Congress on March 14, 1997. Consequently, if DOD intends to include
these policies and procedures in the framework, we believe it should
limit making computer center decisions and investments to those that
meet critical technology needs to operate the centers until the
framework is finalized.
In commenting orally on this report, OMB stated that it believed our
report overemphasized the importance of consolidating mid-tier
processors within the context of OMB Bulletin 96-02. We disagree; we
continue to believe that the consolidation strategy needs to include
mid-tier processors as they are a vital component of the services
offered by the computer centers.
---------------------------------------------------------- Letter :9.1
We are sending copies of this report to the Ranking Minority Member
of your Committee and the Chairmen and Ranking Minority Members of
the House and Senate Committees on Appropriations, the Senate
Committee on Governmental Affairs, the House and Senate Committees on
the Budget, the Senate Committee on Armed Services, and the House
Committee on Government Reform and Oversight. Also, we are sending
copies to the Secretaries of Defense, the Army, the Navy, and the Air
Force; the Department of Defense Chief Information Officer; the
Director of the Defense Information Systems Agency; the Director of
the Defense Logistics Agency; the Director of DISA's Westhem Command;
the Director of Office of Management and Budget; and other interested
parties. Copies will be made available to others upon request.
If you have any questions about this report, please call me at (202)
512-6240 or Mickey McDermott, Assistant Director, at (202) 512-6219.
Other major contributors to this report are listed in appendix IV.
Sincerely yours,
Jack L. Brock, Jr.
Director, Defense Information and
Financial Management Systems
SCOPE AND METHODOLOGY
=========================================================== Appendix I
To assess whether DOD has an effective framework in place for making
and executing its computer center decisions, we interviewed staff and
obtained documentation from the following federal activities:
-- the Office of Management and Budget's Office of Information
Policy and Technology Branch, which has responsibility for
overseeing agency implementation of OMB Bulletin 96-02;
-- the General Service Administration's Office of Governmentwide
Policy and Federal Systems Management Center, which provided
documentation on matters federal agencies should consider when
making consolidation, optimization, or outsourcing decisions;
-- the Office of the Assistant Secretary of Defense, Command,
Control, Communications, and Intelligence, which is the office
of DOD's Chief Information Officer;
-- various offices of the Defense Information Systems Agency,
primarily in Arlington, Virginia; and
-- Army, Navy, and Air Force staffs and offices in Arlington,
Virginia, with responsibility for making decisions on
consolidating, optimizing, or outsourcing their computer center
operations.
We also met with managers from corporations that had successfully
consolidated, modernized, and outsourced their computer centers. We
identified these corporations through discussions with private-sector
consultants and Defense computer center officials. The corporations
contacted were
-- Boeing Computing Service, Belleview, Washington;
-- Electronic Data Systems, Plano, Texas; and
-- GTE Corporation, Fairfax, Virginia.
Through these interviews and related documentation, we analyzed how
these companies strategically direct and oversee their decisions on
alternatives and how they determine the cost and measure the
performance of their computer center operations.
We also met with consultants who advise computer center managers on
improving their services. The consultants contacted were the Center
for Naval Analyses, Compass America, Inc., Coopers and Lybrand, and
the Gartner Group. In these discussions, we identified best
practices and important performance measures that they believe well
managed computer centers should use to benchmark their performance
with other computer centers. In addition, we interviewed senior
officials at the Defense Science Board to discuss the Board's
high-level study done for DOD management on the outsourcing of select
DOD activities, including its computer centers.
Finally, we met with DOD officials in the Office of the Assistant
Secretary of Defense for Command, Control, Communications, and
Intelligence to discuss their actions to implement a departmentwide
decision-making framework for making computer center investment
decisions. To assess the effectiveness of DOD's framework, we
compared the framework with best practices used by leading
organizations and the Clinger-Cohen Act. Also, through this office,
we obtained and analyzed DOD's submissions to OMB in compliance with
OMB Bulletin 96-02 to determine whether these submissions met OMB's
requirements and had been prepared to meet departmentwide information
processing needs. We did not validate the accuracy of the numbers
provided by DOD on its computer centers.
Our work was performed from March 1996 through January 1997 in
accordance with generally accepted government auditing standards. We
performed our work primarily at the office of the Assistant Secretary
of Defense for Command, Control, Communications, and Intelligence and
at DISA headquarters offices in Arlington, Virginia.
(See figure in printed edition.)Appendix II
COMMENTS FROM THE DEPARTMENT OF
DEFENSE
=========================================================== Appendix I
See comment 1.
(See figure in printed edition.)
See comment 2.
(See figure in printed edition.)
See comment 3.
See comment 4.
See comment 5.
(See figure in printed edition.)
Now on p. 19.
(See figure in printed edition.)
Now on p. 19.
(See figure in printed edition.)
The following are GAO's comments on the Department of Defense's
letter dated March 10, 1997.
GAO COMMENTS
1.We acknowledge that DOD has reported significant savings through
its prior consolidation efforts and have expanded the report to
reflect the fact that DOD has consolidated 194 DOD computer centers
into 16 DISA megacenters, at a reported reduction in processing costs
of over $500 million. (See section entitled, DOD Recognizes Benefits
of Further Consolidating, Modernizing and Outsourcing Computer
Centers.) As appropriate, we also expanded the report to acknowledge
DOD's use of industry practices to help make these reductions. (See
section entitled, DOD Lacks Critical Decision-making Tools for
Consolidation Efforts.)
2.We agree that DOD needs to determine whether further economies and
efficiencies are possible and, if so, what strategies should be
employed to reap these savings. The recommendations to DOD and OMB
contained in this report are intended to facilitate and guide these
determinations.
3.The report fully describes the differing views of DOD and OMB
officials for interpreting OMB Bulletin 96-02 in two broad areas:
(1) DOD's consolidation plans for its non-mainframe small and
mid-tier computer centers and (2) the number of DOD plan submissions
required by OMB. In the report, we pointed out that OMB officials
did not agree with DOD's interpretation that non-mainframe computer
centers should only be included in their inventories but not their
consolidation strategies. OMB's position, which we support, is that
non-mainframe centers should have been described in DOD's inventories
and consolidation strategies, as the purpose of OMB Bulletin 96-02 is
to look for ways to consolidate all DOD's computer centers, not just
its mainframe computer centers. We made our recommendation that OMB
clarify the Bulletin with regard to its mid-tier consolidation
criteria in order to preclude any future confusion.
We further recommended that OMB clarify in the Bulletin that while
DOD has previously been permitted to provide separate submissions for
the three services and for DOD, it should be required to provide a
single, integrated submission for the entire Department.
4.We provided and discussed an earlier draft of this report with DOD
officials and have incorporated their comments as appropriate to
improve the accuracy of the report. The reference in DOD's letter to
our handling of the Army's centers in table 2 refers to wording that
was provided by the Army. However, note b to table 2 has been
expanded to reflect Army's views that some of its centers provide
unique missions (for example, command and control, and National
Guard).
5.We did not include DOD's second enclosure in appendix II because it
is an annotated copy of this report. This enclosure contained a few
technical comments, which we have incorporated into the final report.
DETAILED INFORMATION ABOUT
COMPUTER CENTERS
========================================================= Appendix III
This appendix provides information on the numbers of mainframes and
mid-tier and small processors owned by the services and components.
Table III.1
Number of Mainframes and Mid-tier and
Small Processors Owned by the Services
and Components
Mid-tier/
Service/component Mainframes Small
---------------------------------------- ------------- -------------
DISA 162 173
Air Force 22\a 184
Army 79\b 378
Navy 28 107
Defense Commissary Agency 0 20
Defense Intelligence Agency 5 82
Defense Investigative Service 1 4
Defense Logistics Agency 19 84
National Imagery and Mapping Agency 5\c 0
Defense Special Weapons Agency 1 27
======================================================================
Total 322 1,059
----------------------------------------------------------------------
\a Includes the exempted computers.
\b Only 11 Army mainframes are being considered for consolidation.
Remaining mainframes will be replaced with minicomputers to support
unique local missions, such as civil works, command and control,
research and development, etc.
\c Only one mainframe, with four remote-entry machines.
Source: Agency and component submissions to OMB.
Table III.2
MIPS (Mainframe Operating Capacity) of
the Individual Services and Components
Service/component MIPS
------------------------------------------------------------ --------
DISA 7,882
Air Force 1,176
Army 4,692\a
Navy 195
Defense Commissary Agency \b
Defense Intelligence Agency 908
Defense Investigative Service 43
Defense Logistics Agency 1,007
National Imagery and Mapping Agency 22
Defense Special Weapons Agency 28
======================================================================
Total 15,953
----------------------------------------------------------------------
\a Includes minicomputers or scientific computing; remaining centers
support local or unique missions.
\b The Defense Commissary Agency does not have any mainframes.
Source: Agency and component submissions to OMB.
ANALYSIS OF DOD SERVICE AND
COMPONENT RESPONSES TO OMB
========================================================== Appendix IV
The tables that follow summarize our analysis of the extent to which
DOD services and components complied with the planning elements
called for by OMB.
Table IV.1
DISA's Consolidation Strategy
OMB requirement Agency response to OMB
---------------------------------------- ----------------------------
Alternatives analysis reflecting the DISA did not submit an
technical feasibility and cost- alternatives analysis to
effectiveness of alternatives-- OMB. However, during our
including outsourcing. review, we found that DISA
had analyzed the costs and
benefits associated with
(1) outsourcing megacenter
services and
(2) consolidating 16
megacenters into 6 centers.
Architecture design, or technical Technical architecture
solution, based on selected data center submitted to OMB was not
consolidation alternative, and based on an approved
identifying the receiving and closing alternative analysis, nor
data centers and workload realignment as did it identify receiving
well as the communications architecture. and closing data centers.
Architecture did not address
workload realignment or
communications architecture.
High-level implementation approach No implementation approach
identifying major consolidation tasks submitted to OMB.
and presenting a schedule, milestones,
and resources.
Funding plan identifying and forecasting No funding plan submitted to
costs associated with the consolidation OMB.
process and funding requirements for all
major tasks associated with the
consolidation.
Exceptions that could not be included in No exceptions identified to
the consolidation plan. OMB.
----------------------------------------------------------------------
Source: DISA consolidation strategy, dated June 6, 1996.
Table IV.2
Air Force's Consolidation Strategy
OMB requirement Agency response to OMB
---------------------------------------- ----------------------------
Alternatives analysis reflecting the The Air Force did not submit
technical feasibility and cost- an alternatives analysis to
effectiveness of alternatives-- OMB; however, it did
including outsourcing. describe plans to move
towards a mid-tier
architecture for some of its
centers and to outsource
those centers that cannot be
moved to mid-tiers.
Architecture design, or technical The Air Force did not submit
solution, based on selected data center an architectural design to
consolidation alternative, and OMB, only individual
identifying the receiving and closing computer center approaches
data centers and workload realignment as to consolidation.
well as the communications architecture.
High-level implementation approach A partial implementation
identifying major consolidation tasks approach was submitted to
and presenting a schedule, milestones, OMB in that schedules and
and resources. milestones were provided.
Funding plan identifying and forecasting No funding plan submitted to
costs associated with the consolidation OMB, although some data were
process and funding requirements for all provided on estimated
major tasks associated with the savings from consolidation.
consolidation.
Exceptions that could not be included in Exceptions were requested
the consolidation plan. from OMB for the following
reasons: 4 centers performed
applications programming, 12
centers operated national
security systems, 1 center
met OMB's exception criteria
by having less than five
full-time employees, and 2
centers were transitioning
to the Air Force Working
Capital Fund.
----------------------------------------------------------------------
Source: Air Force consolidation strategy, dated September 18, 1996.
Table IV.3
Army's Consolidation Strategy
OMB requirement Agency response to OMB
---------------------------------------- ----------------------------
Alternatives analysis reflecting the Alternatives analyses were
technical feasibility and cost- submitted to OMB for the
effectiveness of alternatives-- computer centers Army
including outsourcing. considered candidates for
consolidation or
outsourcing: the separate
Single Agency Manager
computer centers operated by
the Air Force and the Army
and two Army personnel
computer centers.
Outsourcing to DISA and
leasing were among the
alternatives considered.
Architecture design, or technical Architectural designs were
solution, based on selected data center submitted for the Single
consolidation alternative, and Agency Manager and for the
identifying the receiving and closing two personnel computer
data centers and workload realignment as centers.
well as the communications architecture.
High-level implementation approach A high-level implementation
identifying major consolidation tasks approach was submitted to
and presenting a schedule, milestones, OMB. Resources, but not
and resources. schedules and milestones,
were submitted for major
consolidation tasks.
Funding plan identifying and forecasting A funding plan was not
costs associated with the consolidation submitted, but Army plans to
process and funding requirements for all chargeback costs to
major tasks associated with the customers.
consolidation.
Exceptions that could not be included in Exceptions identified for
the consolidation plan. areas such as
National Guard, civil works,
intelligence, command and
control, research and
development, and wargaming.
Centers not analyzed for
consolidation were reported
as centers that support
networks or systems in a
"distributed environment."
----------------------------------------------------------------------
Source: Army consolidation strategies, dated May 3, 1996, and August
2, 1996; further analyses provided for the Single Agency Manager and
the personnel centers in July 1996.
Table IV.4
Navy's Consolidation Strategy
OMB requirement Agency response to OMB
---------------------------------------- ----------------------------
Alternatives analysis reflecting the Each of 10 Navy commands
technical feasibility and cost- reported a consolidation
effectiveness of alternatives-- strategy for its centers
including outsourcing. that Navy believed met OMB's
criteria for consolidation.
Two of these commands, the
Bureau of Naval Personnel
and the Naval Supply Systems
Command, reported that DISA
megacenters already
processed their information.
The Naval Air Systems
Command and the Navy
Facilities Engineering
Command also described plans
for DISA to process their
information. Two other
commands, the Naval Sea
Systems Command and the
Bureau of Medicine and
Surgery, provided
alternative analyses
supporting their decision to
continue to process their
information in-house. The
remaining four commands did
not provide alternative
analyses to OMB.\a
Architecture design, or technical Architectural designs were
solution, based on selected data center submitted but were not based
consolidation alternative, and on alternatives analyses for
identifying the receiving and closing four commands.
data centers and workload realignment as
well as the communications architecture.
High-level implementation approach Completion dates were
identifying major consolidation tasks provided, but schedules of
and presenting a schedule, milestones, major consolidation tasks or
and resources. resource needs were not
provided.
Funding plan identifying and forecasting A funding plan was not
costs associated with the consolidation provided to OMB, but Navy
process and funding requirements for all plans to fund its
major tasks associated with the modernization efforts
consolidation. through its information
technology budget.
Exceptions that could not be included in Not applicable.
the consolidation plan.
----------------------------------------------------------------------
\a In subsequent discussions, a Navy official explained that an
eleventh command, the Atlantic Fleet, operated four of the Navy's
computer centers reported in table 2. This command's plan to
transfer its centers--all on the Base Realignment and Closure (BRAC)
list--to DISA had inadvertently been omitted from the Navy's
strategy. Also omitted was the Navy Sea System Command's plans to
transfer the processing of its Dahlgren Center to DISA in fiscal year
1998.
Source: Navy consolidation strategy, dated May 22, 1996; addendum
provided on August 8, 1996.
Table IV.5
Defense Commissary Agency's
Consolidation Strategy
OMB requirement Agency response to OMB
---------------------------- ----------------------------------------
Alternatives analysis The agency has decided to move to a
reflecting the technical client/server architecture. No
feasibility and cost- consolidation strategy was submitted for
effectiveness of this move.\a
alternatives--including
outsourcing.
Architecture design, or None submitted.
technical solution, based on
selected data center
consolidation alternative,
and identifying the
receiving and closing data
centers and workload
realignment as well as the
communications architecture.
High-level implementation None submitted.
approach identifying major
consolidation tasks and
presenting a schedule,
milestones, and resources.
Funding plan identifying and None submitted.
forecasting costs associated
with the consolidation
process and funding
requirements for all major
tasks associated with the
consolidation.
Exceptions that could not be No exemptions requested.
included in the
consolidation plan.
----------------------------------------------------------------------
\a According to DOD officials, the Defense Commissary Agency did not
believe it had to submit a consolidation strategy because it only has
mid-tier computers.
Source: Defense Commissary Agency consolidation strategy, dated
February 14, 1996; mission needs statement for the Agency's Point of
Sale Modernization Program, dated May 6, 1994.
Table IV.6
Defense Intelligence Agency's
Consolidation Strategy
OMB requirement Agency response to OMB
---------------------------- ----------------------------------------
Alternatives analysis The agency reported that it met the OMB
reflecting the technical target MIPS for a minimum size computer
feasibility and cost- center at both of its computer centers.
effectiveness of
alternatives--including
outsourcing.
Architecture design, or See above.
technical solution, based on
selected data center
consolidation alternative,
and identifying the
receiving and closing data
centers and workload
realignment as well as the
communications architecture.
High-level implementation See above.
approach identifying major
consolidation tasks and
presenting a schedule,
milestones, and resources.
Funding plan identifying and See above.
forecasting costs associated
with the consolidation
process and funding
requirements for all major
tasks associated with the
consolidation.
Exceptions that could not be See above.
included in the
consolidation plan.
----------------------------------------------------------------------
Source: Defense Intelligence Agency consolidation strategy, dated
May 7, 1996.
Table IV.7
Defense Investigative Service's
Consolidation Strategy
OMB requirement Agency response to OMB
---------------------------- ----------------------------------------
Alternatives analysis Although the agency meets the minimum
reflecting the technical target size criteria, it still provided
feasibility and cost- alternatives analyses supporting (1)
effectiveness of move from traditional data center to
alternatives--including client/server architecture and (2)
outsourcing. continuing to process in-house
Architecture design, or Not applicable.
technical solution, based on
selected data center
consolidation alternative,
and identifying the
receiving and closing data
centers and workload
realignment as well as the
communications architecture.
High-level implementation Not applicable.
approach identifying major
consolidation tasks and
presenting a schedule,
milestones, and resources.
Funding plan identifying and Not applicable.
forecasting costs associated
with the consolidation
process and funding
requirements for all major
tasks associated with the
consolidation.
Exceptions that could not be Requested exemption because the
included in the Service's computer center meets the
consolidation plan. minimum target size for computer
centers.
----------------------------------------------------------------------
Source: Defense Investigative Service consolidation strategy, dated
May 30, 1996, and Implementation Plan dated August 31, 1996.
Table IV.8
Defense Logistics Agency's Consolidation
Strategy
OMB requirement Agency response to OMB
---------------------------- ----------------------------------------
Alternatives analysis Alternatives analyses were submitted for
reflecting the technical 3 of the agency's 12 computer centers,
feasibility and cost- with selection of option to consolidate
effectiveness of in-house operations. Two of the agency's
alternatives--including computer centers were exempt (see below)
outsourcing. and another two were designated for
BRAC. Alternative analyses were not
submitted for the remaining 5 computer
centers, which will support the agency's
planned Distribution Standard System.
Architecture design, or Not submitted for centers that will
technical solution, based on support the Distribution Standard System
selected data center because the agency believed these
consolidation alternative, requirements were not applicable to its
and identifying the overall strategy.
receiving and closing data
centers and workload
realignment as well as the
communications architecture.
High-level implementation Not submitted for centers that will
approach identifying major support the Distribution Standard System
consolidation tasks and because the agency believed these
presenting a schedule, requirements were not applicable to its
milestones, and resources. overall strategy.
Funding plan identifying and Not submitted for centers that will
forecasting costs associated support the Distribution Standard System
with the consolidation because the agency believed these
process and funding requirements were not applicable to its
requirements for all major overall strategy.
tasks associated with the
consolidation.
Exceptions that could not be Two of the agency's computer centers
included in the meet OMB's target size for computer
consolidation plan. centers.
----------------------------------------------------------------------
Source: Defense Logistics Agency's consolidation strategy, dated
July 1, 1996.
Table IV.9
National Imagery and Mapping Agency's
Consolidation Strategy
OMB requirement Agency response to OMB
---------------------------- ----------------------------------------
Alternatives analysis Alternatives analyses were not
reflecting the technical submitted, but the agency plans to
feasibility and cost- transition processing from remaining
effectiveness of agency computer centers to DISA
alternatives--including megacenters by the middle of fiscal year
outsourcing. 1998.
Architecture design, or See above.
technical solution, based on
selected data center
consolidation alternative,
and identifying the
receiving and closing data
centers and workload
realignment as well as the
communications architecture.
High-level implementation Broad milestones for transition were
approach identifying major provided, but no additional information.
consolidation tasks and
presenting a schedule,
milestones, and resources.
Funding plan identifying and A funding plan was not provided to OMB.
forecasting costs associated The agency plans to provide OMB with a
with the consolidation funding plan if DISA can process its
process and funding applications.
requirements for all major
tasks associated with the
consolidation.
Exceptions that could not be None requested.
included in the
consolidation plan.
----------------------------------------------------------------------
Source: National Imagery and Mapping Agency's consolidation
strategy, dated May 3, 1996, and the agency's implementation plan
(not dated).
Table IV.10
Defense Special Weapons Agency's
Consolidation Strategy
OMB requirement Agency response to OMB
---------------------------- ----------------------------------------
Alternatives analysis Agency conducted an alternatives
reflecting the technical analysis to support the consolidation of
feasibility and cost- its information processing in-house.
effectiveness of Alternatives such as commercial
alternatives--including outsourcing were not considered because
outsourcing. of the agency's command, mission, and
security considerations.
Architecture design, or A complete architectural design was
technical solution, based on provided.
selected data center
consolidation alternative,
and identifying the
receiving and closing data
centers and workload
realignment as well as the
communications architecture.
High-level implementation A complete high-level response was
approach identifying major provided.
consolidation tasks and
presenting a schedule,
milestones, and resources.
Funding plan identifying and A complete funding plan was provided.
forecasting costs associated
with the consolidation
process and funding
requirements for all major
tasks associated with the
consolidation.
Exceptions that could not be None requested.
included in the
consolidation plan.
----------------------------------------------------------------------
Source: Defense Special Weapons Agency's consolidation strategy,
dated August 16, 1996.
MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix V
ACCOUNTING AND INFORMATION
MANAGEMENT DIVISION, WASHINGTON,
D.C.
William D. Hadesty, Technical Director
Robert C. Sorgen, Evaluator-in-Charge
Danny R. Latta, Technical Advisor
Cristina T. Chaplain, Communications Analyst
OFFICE OF THE GENERAL COUNSEL
Frank Maguire, Senior Attorney
OFFICE OF THE CHIEF ECONOMIST
Harold J. Brumm, Jr., Senior Economist
KANSAS CITY REGIONAL OFFICE
Karl G. Neybert, Evaluator
*** End of document. ***