Thing

Defense Logistics Agency’s Year 2000 Program
Managing Organization-Wide Conversion and Compliance Sarah J. Reed
Defense Logistics Agency System Design Center
The Defense Logistics Agency considers the year 2000 (Y2K) problem mission-critical, and we have treated itas such in planning and executing the largest maintenance effort we have undertaken. The agency kicked offa formal Y2K project in November 1995 with nearly a full year of planning, preparation, and piloting. Thisarticle discusses our Y2K initiative and our experiences in raising awareness and in assessing, renovating, andvalidating our systems. Our program, built on available industry research and our experiences, has beenmodified as we have gained fresh insight and assimilated new lessons learned. The Defense Logistics Agency interview and survey. Responses were developing funding proposals, conver- consistency, then queried and analyzed.
and tools to facilitate analysis, renova- tion, and testing. Conversion progress is assessed on a regular basis through status date references are high impact; that is, sulted in a fairly comprehensive applica- automated information systems (AIS).
Expected to be rehosted,
for-service activity, and funded as we are reengineered, or replaced but not
Our Application Profile
compliant (16 percent, 19 applica-
tions). If replacement initiatives experi-
ponent data provides an idea of our size: across multiple system structures. The Component
Total number
systems has been for us, unprecedented.
replacement system initiatives experience System Impact Assessment
Findings, Recommendations,
and Resulting Actions
tions about date practices, levels of cus- Expected to remain in place be-
yond 2000 and compliant (33.6 per-
cent, 40 applications). Systems thought
Expected to remain in place be-
yond 2000, but not compliant (47.1
less risk than those that were not compli- percent, 56 applications). We ad-
CROSSTALK The Journal of Defense Software Engineering 11
Year 2000
worked to establish a liaison with ourchief deployment site.
Vendor Product Risk
Related Observations and
Date formats. The wide variety of
Assessment and Mitigation
Recommendations
AISs not included in the survey data.
points out the need for greater standard- survey for a variety of reasons, e.g., they always practical for large, complex legacy straints associated with them. Because of • Enter response data into a repository.
Database Management Systems
• Analyze for risk classification.
(DBMS). The 15 different database
Vendor product compliance. The
discovered that having a date field with a gram office continues to pursue the issue publish its findings along with appropri- Awareness. In spite of briefs to senior
Interface recommendations. Initially,
Identification of Development
Environment Software
the initiative in January 1997. We subse- rums to facilitate the sharing of progress, the system and filter building to protect Assessment process. We knew going
into the assessment process that develop- ing data—and not on self-initiated inter- expected to find a similar identification agree that an interface strategy was nec- analyzing the results would be a difficult not to change the interface formats until statistical analysis or survey science back- were not previously identified by the application support areas or the acquisi- strategy was delivered in August 1997.
the quality of the survey responses, often tion office. We established possible rea- Release Management. AIS upgrades
are normally released to multiple deploy- sented the additional challenge of simul- Where Are We Now?
12 CROSSTALK The Journal of Defense Software Engineering
Defense Logistics Agency’s Year 2000 Program: Managing Organization-Wide Conversion and Compliance Follow up with vendors expected
to provide compliant products at a
future date. For these products, we
dor to obtain the required information.
Develop testing strategies for those
products vendors have stated are com-
Improve software portfolio man-
their application. Upgrade strategies will pliant. Ascertaining the state of vendor
agement processes. The great discrep-
timely arrival, installation, and testing of Develop independent strategy for
IBM products. A planned operating
published studies still leaves a vulnerabil- Table 1. Y2K support tool categories according to DLA internal priorities. f rt via the automated generation of test data r ss-centur ,y and next century test cases.
advance dates in test data transactions.
Ability to simulate dates other than cu e o r priority; our strategy utilizes expansion on limited basis CROSSTALK The Journal of Defense Software Engineering 13
Year 2000
in Table 1. The higher priority tools (1, Review and Selection of
2, 3, etc.) are those tools for which little sometimes difficult to get it installed.
Support Tools
or no existing internal capability existed Tool Recommendations and
Implementations
tify potential tools and determine appro- Certification Process to
Benchmark and Report System
Compliance
dural logic or date windowing solution.
Lessons Learned
Table 2. Characteristics of six DLA Y2K conversion projects. f r UNIX shel , C modifications. In-house- 14 CROSSTALK The Journal of Defense Software Engineering
Defense Logistics Agency’s Year 2000 Program: Managing Organization-Wide Conversion and Compliance Renovation. The selected approach
different certification levels reflecting Certification Process
approach is the only risk viable approach to achieving Y2K compliance in time.
Awareness and Communications. A
certification guidance and a checklist.
to presentation to the customer for final Pilot Project Study
the user-interface layer. It is important for the user to understand the impacts of Lessons Learned
various system, interface, and user inter- • In spite of our efforts to streamline it, • Identification of not-yet-anticipated ure risk to the existing system. Addition- ally, intention to replace a legacy system does not entirely relieve the customer of tions and additional testing and certifi- Pilot Project Profile
cation activities may still be required to Table 2 summarizes the characteristics of been, and continues to be, difficult.
Lessons Learned
Estimation of the Effort Involved.
effectively used by all pilots. To compen- pilot project lines of code (LOC) costs.
Tools. In-house tool capability was
• In spite of our best efforts to baseline CROSSTALK The Journal of Defense Software Engineering 15
Year 2000
• Focus on the underlying hardware during pilot testing efforts. This realiza- tion from pilot efforts escalated our tool • Obviously, remediation for systems assessment and procurement initiatives.
Testing. Testing time and effort was
planning within the system effort, Provide Sufficient Investment in
even as the system is being renovated.
• Carving out enough time for inte- tools and training, without effective test cient to ensure success of the effort.
in testing procedures within the organiza- tion affected attempts to baseline a certifi- expertise to guide the test plan develop- cation process. Effort expended in defin- About the Author
ing Y2K-specific testing practices and in Sarah J. Reed began
strategy for specific needs related to test Be Prepared for High-Level
environments, test tools, test utilities, Visibility
We recognize the benefit of high visibil- ity as the crucial components of sponsor- the Organizational Culture Initiative, and Release Management. Close coordi-
tool releases but also for a larger-than- she led the effort to establish a baseline leases into the field. Where file conver- these calls for information. Lack of such Observations and Conclusions
sons learned as they pertain to significant have found to be the most critical factors aspects of our effort. Here we offer some overall observations and conclusions.
• Sustaining organizational commit- Provide a Sufficient Resource
• Understanding the organizational Voice: 614-692-9296 DSN 850-9296Fax: 614-692-8393 DSN 850-8393 Investment
baseline at the outset of the initiative.
• Understanding the challenge in ap- • The adequacy of testing tools and 16 CROSSTALK The Journal of Defense Software Engineering

Source: http://www.crosstalkonline.org/storage/issue-archives/1998/199801/199801-Reed.pdf