FY 1981 REPORT ON ADP IN THE CIA (Sanitized)

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP84-00933R000100080001-4
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 17
• Document Creation Date: December 15, 2016
• Document Release Date: August 6, 2003
• Sequence Number: 1
• Publication Date: May 18, 1981
• Content Type: MF

ZOO
Approved For Releas 6 : CIA-RDP84-009338000100080001-4
ODP-81-609
18 MAY 1981
MEMORANDUM FOR: Executive Committee Members
VIA Deputy Director for-Administration
FROM Bruce T. Johnson
Director of Data Processing
SUBJECT FY 1981 Report on ADP in the CIA
REFERENCE Memorandum to DDCI from D/ODP dtd.
16 May 1980, Subject: Fiscal Year 1980
Review of Automatic Data Processing (ADP)
Projects (Exec. Registry 9037-80/2)
Summary: This memorandum is a substitute for
the annual EXCOM briefing on ADP in CIA. It
concentrates-on the services provided by the
Office of Data Processing (ODP) and concludes
that ODP's existing and planned capabilities
are basically adequate to meet the foreseeable
central computer service requirements of
CIA. Some new investments to ensure
reliability and availability will be
necessary, and an increase in applications
development manpower seems called for. 0
1. Introduction: In 1977 the management of CIA's ADP
resources was studied by the Executive Advisory Group (EAG), the
forerunner of today's Executive Committee. Out of those studies
developed the practice, beginning in 1978, of having the Director
of Data Processing present a "State of ODP" briefing to the
Committee, followed by reports from user components on major ADP
activities, which were expected to cost $250,000 in the current
fiscal year for ongoing projects, or th
a
t same amount over
three
years for new initiatives.
2.
In
fiscal years 1978,
1979, an
d
1980, late in the
first
quarter
or
early in the second
quarter,
the EXCOM devoted
several
meetings
to
the review of ADP,
beginnin
g
each time with an
OI)P
25X1
25X1
25X1
Approved For Release 2qF6TCIA-RD~84-00933R000100080001-4
 Approved For Release 20 1 1i1ETCIA-RDP84-00933R000100080001-4
overview based upon data collection by ODP from major users of
ADP. Following each annual review, a guidance memorandum was
issued, validating the proposed use of ADP resources and
establishing limits on the growth of individual projects.
Subsequently, if rates of consumption of these resources
increased enough to cause a project to exceed its authorized
limit by more than 10%, the component was required to notify the
Comptroller and submit a justification for the increase. After
the 1980 review, at the request of the DDCI, a summary of the
results of the review was prepared (see reference).
D
3. Subsequent to the 1980 review the Information Handling
Task Force completed its work and, as a result of one of its
recommendations, the position of Information Handling Systems
Architect was established. In approving a tentative charter for
the Architect, the EXCOM included the responsibility to report on
the "state of Information Handling to EXCOM," adding
parenthetically "(incorporates ADP review)." II
4. As FY 1981 began, ODP once more collected the data
needed for its part of the annual review, although the timing and
nature of the review were uncertain, given the arrival on the
scene of the new Architect and the interruption in the routine
work of thy? EXCOM and its staff caused by the transition to new
Agency management. Discussions involving the EXCOM staff, the
Architect, and ODP led to the decision to submit a brief surnmary
report as a substitute for the briefings previously provided.
That report follows.
0
25X1
25X1
5. Agency 'ADP Budget: Attachment A represents the Agency
ADP budget for fiscal years 1980, 1981, and 1982, by Directorate,
with separate entries for SAFE, NPIC and ODP, which together
account for 84% of the 1981 total. In FY 1981 the Agency ADP
budget increased 18% over FY 1980, rising from 25X1
principally due to the scheduled rise in the SAFE funding
profile, although imagery-related budgets also rose and of course
inflation played its part. It is estimated that in this fiscal
year Agency staff personnel will devote to 25X1
ADP-related activities, with the DDA accounting for 61% of the
effort. The ADP budget for FY 1982 is 35% larger than FY 1981,
the increase attributable to the upgrade of the NPIC Data System
(approximately and CAMS 2
plus a few smaller projects like 4C and, of course,
inflation.
6. Consumption of ODP Resources in FY 1980: Each month ODP
provides its customer components with a Project Activity Report
(PAR) which shows how much of ODP's resources were consumed by
t ', 2
Approved For Release 2Q0
tTA-RDP84-00933R000100080001-4
 SLUR
Approved For Release 2003/08/26 : CIA-RDP84-00933R000100080001-4
each of the component's ADP projects. The pricing system is
designed to permit nominal "recovery" of the entire ODP budget.
The primary charges, accounting for about 65% of the total, are
for the use of central processing services. The other charges
included direct contract costs (15%), application software
development and maintenance (12%) and dedicated support such as
computer systems, terminals and direct access storage space
7. ODP Supported Major Projects by Directorate: ODP
supported 17 major ADP projects in FY 1981. Attachment B shows.
the level of investment in these projects by Directorate, and the
percentage of increases expected in their use of ODP's major
services, Batch, Interactive (VM), and Database (GIMS).
Attachment C lists each of these 17 major ADP projects, which
range from 1 to 18% of the OD' budget aAd together consume 57% of
that budget. (We also received reports on 17 other projects
which did not meet the $250K threshold requirement but which were
significant enough to include in the annual review. The projects
in this second category account for an additional 6% of ODP's
resources.)
D
8. Major Component-Budgeted ADP Initiatives: Attachment D
provides directorate-level summary information for FY 1981 and
1982 on 14 component-budgeted ADP initiatives, and Attachment. E
identifies the individual projects. Although primarily funded by
the customer component, many of these projects entail some level
of ODP support, as shown in the tables. Most of these projects
will continue into FY 1983, and resources to support them have
been included in the Program Plans submitted by the sponsoring
components. (Because SAFE is the subject of separate review by
its Steering Committee and the Directors of DIA and CIA, it has
been excluded from previous EXCOM reviews, and its budget has not
been included in Attachments D or E. The ODP budget for the
support of SAFE development is included in Attachment C,
however.)
D
9. Review of Major Projects: Our review of the data on
these major projects reveals several issues worth highlighting.
Approved For Release 200 RDP84-009331000100080001, 4
 ~U(i hU l
Approved For Release 2003/08/26 : CIA-RDP84-00933R000100080001-4
a. Community-related Activities. The COMIREX
Automated Management System. (CAMS) continues
to be the largest of the ODP-supported
ts. Total cost of services represents
I lof the ODP budget. More significant,
however, is the continuing CAMS requirement
for of the staff employees in the
applications component of ODP of the
total applications staff when one counts
contractor personnel).
b. Interface to Central Data Banks. Several
component-budgeted projects involve
minicomputer-based systems for specialized
processing needs. Interfaces to central ODP
files are required for access to pertinent
data. This relatively new requirement for
delivery of large amounts of data in
electrical form to remote processing sites
further heightens our already growing concern
over the availability and reliability of data
processing and communications facilities to
support the rapidly increasing number of users
who require frequent and immediate access to
their data. New investments will have to be
made to ensure that our network of computer
and communication systems operate with the
necessary availability and reliability to
guarantee the continuing delivery of critical
services to our end users.
c.
Terminal
Availability.
The tight limits
imposed
continue
on ODP for new
to provide a
terminal
constraint
acquisitions
on the growth
of project activity.
Budgetary
restrictions
have hampered our ability to provide anything
beyond the minimum level program requirements
of our end users. By FY 1983, we will only be
able to fund 77% of our user's minimum level
requirements and less than 25% of their
overall requirements. At least two
directorates (DDO and NFAC) are planning to
fund their most critical requirements that
cannot be supported by ODP.
4
Approved For Release 2003/ ? RDP84-009338000100080001-4
25X1
 Approved For Release 2003/08/26 : CIA-RDP84-00933R000100080001-4
d. Applications Development Resources. The
success of certain projects reviewed is very
much dependent upon the availability of
adequate program development resources.
Because of the increasing maintenance burden
we have had to bear as new systems are added
to the existing base of developed production
systems, the present share of ODP resources
available for new applications development has
been shrinking drastically at the same time
that the requirement for new application
development from virtually all parts of the
Agency has reached an unprecedented level (the
backlog exceeds 50 workyears). Although our
Program Plan for FY 1983 contains 35
additional staff positions to help meet this
need, our problems are more immediate. The
schedules for several of the ODP-supported
projects (in particular, PAYROLL) are in
jeopardy as a result.
0
10. ODP Workload Trends: The major projects provide a basis
for projecting only part of the demand for ODP's services. The
trend lines for services to all ojects, large and small, are
all rising. We measure these trends for Batch Service (average
prime-time as well as average peak 24-hour workload), Interactive
(VM) Services, Database Management Services (GIMS), DDO Services
and CAMS Service.
a. Batch:? Still the most economical way to meet
the needs of a large number of users, Batch
continues to grow at the rate of 10 to 15%
annually. To avoid driving customers to the
use of more expensive on-line service, we size
our Batch capability to provide reasonable
prime-time turnaround, with assurance of
24-hour-turnaround capacity for lower priority
work. When any other service is forced into
its backup configuration, Batch is usually the
first to be reduced, but because it can be
divided among several CPU's the impact on
users can be cushioned. Batch capacity is
effectively augmented whenever other demands
cause us to expand our total computing power,
and plans for other services ensure adequate
capacity here.
25X1
Approved For Release 2003/O :ftrDP84-00933R000i00080001-4
 SLUhEI
Approved For Release 2003/08/26 :CIA-RDP84-00933R000100080001-4
b. Interactive: VM service, upgraded in FY 1980
to an Amdahl V8, continues to grow and a
larger IBM 3033 multiprocessor (MP) was
installed in March 1981. Currently, there are
users and the issuance of passwords for
new users continues to increase steadily. In
1983 we project a workload which will exceed
the capacity of the IBM 3033-MP. Future
expansion will require ODP to examine
alternative means of providing support to VM
users. The DO already has a copy of the VM
system available in the Special Center which
affords protection of their data.
c. Database Management Systems: The Generalized
Information Management System (GIMS) is used
to provide central Database Management System
(DBMS) services. Currently, there are some 35
distinct databases on the single GIMS CPU,
with 18 more scheduled for implementation in
FY 1981. Small increases in CPU power are
projected in FY 1981 and FY 1982 to meet a 29%
rate of increase in demand. This service is
also available as-a separate service in the-,I
Special Center.
d. DDO Services: The Directorate of Operations
has provided ODP with a comprehensive set of
near- and long-term requirements. The most
significant of these, in terms of their impact
on ODP's computer systems plans, include (1)
the DDO's requirement for compartmented VM and
GIMS services, and (2) its, requirement for a
single system capable of running all of its
critical on-line applications. This second
requirement is to enable effective operation
of these applications even when the DO second
machine has been preempted to backup CAMS
service. Plans to meet these requirements
call for upgrading one of the two DDO machines
in FY 1981 and the other in FY 1982. Both of
these upgrades will be accomplished by moving
larger CPU's from the Ruffing Computer Center
to the Special Center.
e. CAMS: The COMIREX Automated Management System
(CAMS) will grow because of its direct support
to overhead reconnaissance programs. The
Approved For Release 2003108666: C1~4-RDP84-00933R000100080001-4
 Approved For Release 2003/08/26 : CIA-RDP84-00933R000100080001-4
current CAMS system is expected to satisfy
user requirements until 1984 when major new
collection systems come onstream. A follow on
system, CAMS 2, will then'be required. A
large CPU will be obtained for development of
this new system and will then become the
production system for CAMS 2 and also support
graphics capabilities using minicomputers as
interfaces. II
11. ODP Plans and Investments: The major thrust of ODP
planning and investment through FY 1983 is focused on ensuring
that Batch, Interactive ()IM), Database, DDO services, and CAMS
have the necessary capacity, responsiveness, and availability to
meet growing Agency needs.
a. FY 1981: Two of our basic services will
require upgrading in FY 1981 to meet increased
workloads. The IBM 3033 MP system was
installed to meet the need for more capacity
on our VM interactive system. This service
had reached saturation with 0 simultaneous
user6lonline. Thr-,multiprocessor will allow
us to handle over II simultaneous users with
satisfactory response time. An existing
Amdahl V6 will be freed from its present use
to replace the computer system supporting
critical DDO services. Because of the use of
the same hardware architecture for all ODP
general services, it is possible to upgrade
multiple services with the procurement of a
single new computer system. More capacity
will be provided as a result to handle the 18
new GIMS applications planned for FY 1981.
b. FY 1982:` In FY 1982 a new computer system
will be installed to meet the planned growth
in batch processing requirements. This
acquisition will allow us to also upgrade the
backup capability for both DDO services and
CAMS needs. A second computer will be
installed for the development needs of the
CAMS 2 project. A modest upgrade in database
services will also be provided. In order to
meet the continuously growing requirement for
online storage, we will embark on a major
Approved For Release 20031
hElf-RDP84-OQ933R000100080001-4
25X1
 SEURLS
Approved For Release 2003/08/26 : CIA-RDP84-00933R000100080001-4
upgrade of our DASD (Direct Access Storage
Devices) capacity starting in FY 1982 when
technologically superior DASD will be
available at substantially lower cost than
current technology.
C. FY 1983: Our planning for FY 1983 includes
one new processor to meet the estimated
interactive workload of over II simultaneous
users on the VM system. Other than VM, no
additional processing upgrades are anticipated
in FY 1983. In the absence of any new
requirement, the FY 1982 upgrades will be
adequate to support our other major services
through the end of FY-1983.
d. FY 1984 and Beyond: Our plans for FY 1984 and
beyond call for annual upgrades to meet
projected workload demands for computer
services. We will continue to reallocate to
other needs equipment that is replaced as a
result of upgrading one service, in line with
our policy of effecting improvements in
multiple services from the procurement of a
single new processor. The adequacy of central
computer service can thus be guaranteed
provided that the ODP budget base is protected
from the impact of inflation.
0
12. Conclusion: Although there are outstanding issues in
the areas of applications development, terminal availability and
system reliability, as noted in paragraph 9 above, it is apparent
that ODP has or will have adequate central computing and data
storage capacity to meet projected demand through FY 1982 and our
Program Plan for FY 1983, if approved, will ensure that we
continue to have the necessary computer power. In view of the
fact that we are.. already well into the third quarter of this
fiscal year, we recommend that we and the requesting components
be authorized to continue with the major projects listed in the
attachments at the level of investment requested, subject to
change as a result of decisions on the 1983 Program. Until
formal action is taken to change the reporting process, ODP will
continue as in the past, alerting customers when overruns appear
likely on ODP-supported activities, and assisting them to re ort
the reasons for increased requirements to the Comptroller.
Attachments: a/s
25X1
25X1
25X1
25X1
Approved For Release 2003/0
~AIRDP84-00933R000100080001-4
 Approved For Release 2003/08/26 : CIA-RDP84-00933R000100080001-4
Next 7 Page(s) In Document Exempt
Approved For Release 2003/08/26 : CIA-RDP84-00933R000100080001-4
 Approved For Release ~O6 #8& ,. l?IU_410933R00010008000174
DD/A Reg. 80-1265
Exec. Reg. 9037-80/2
1.: ~.: ~.^.:?- ;? 1. =r
lb May 1980:
,
-. ~F {' '1-^y, ~,~. - f .~ Y } -1jn ,>. .~' ,.y ..: t' 7r'i ?< y .y. rf. f
'~, ~- .. -']? ; r ss .. $ -
+r w.'i~:+~s
-`'- r ,-~' t ?f ..,,.y .r _ rr , a!'?': r. ':ra?v:,: *Tiic. r..t *..tC +~ -"Y, _ Z-i4i
_ ~~ ' * :.-+s. i c r _ ? ~,. s' r. n .t ~k ~...^"?..i ?~+:. -a...~ i".~" _r ~ 'i r?~ ~.`r
r s ~ r'i- ? -p'
MEMORANDUM FOR:.-'Deputy Director of Central intel13gence
VIA. Deputy Director for. Administration=_--
FROM:
SUBJECT:,
REFERENCE:
Bruce T. Johnson -
Director of Data P-rocessing --
Fiscal Year 1980 Review of Automatic Data
Processing (ADP) Projects 0 -
Your memorandum to EXCOM members, same
subject, dated 18 March 1980 (EXCOM 9037-80)
1. This memorandum responds to reference request
that I summarize the lessons learned during the EXCOM-
review of ADP projects and that I solicit the- views of -
the EXCOM members on how the process could be. improved
next year. 'Paragraph 7 contains several recommendations
25X1 for your consideration. _~ . 7
ons learned -in"the-
2. -.We can summarize the - les
s
FY 1980 Review as follows:;
25X-1
25X1 -
In its FY 80 review the EXCOM received .ADP
Project Decision Forms covering 29 ODP-supported
projects and 11 component-bud eted projects repre
senting a total cost of of - _ 251=
the identified in the FY 80 budget
as ADP money. (These figures exclude 25-X1
for Project SAFE, which is separately reviewed by
its CIA/DIA Steering Committee.) Staff review of these summaries led to requests* for briefings
on ten of these projects (five ODP-supported and
five other) . The briefings, including the-' ODP =
overview, were presented in four meetings of the-
Committee and took a total of five'hours, less
than half the time consumed by the 1979 process.
P_s a result of these reviews all on-going projects.
except accelerated CRAFT were approved at the
FY 80 resource levels requested (see reference)..
Specific comments on individual projects are
25X1
Approved For Release 20 1
 Approved For Release 2003/08/26 CIA-RD084-009338000100080001-4
unnecessary in this suiary; details may be
found in the Project Decision For. (CRAFT
has subsequently been the subject of age grata
25X1
25X1
25X1
will be pocsible if crY ire able to pairs rain ; is
budget base at today's levels, adjusted for -
l Ce _ f
review.) _ r
.The MP-supported projects covered by the-review_
represent =of the resources controlled by ODP:'
.-Prom the overview of ODP presented by the Directo
r
of Data Processing, the #XCOM learned that Customer
depandence on and denand for all forzas
f
o
co:rputer
service (batch, database or GIF!.S, and interactive
ar WI) continues to rise, and customer projections-
Of support needed for the activities under review
show that the growth curves for these services are
not likely to change signif ictntly. Plans for
meeting this growing demand were outli.netii in general
terms. Technology is our ally in this a 1i ply/de-aertd
struggle and it Jr, believed that given the drop in
the unit cost of computing, the Office of Data
Processing will be able to keep ahead of d