Tackling a Complex Mainframe to Cloud Migration for HUD
Background
Over the past few decades, a growing challenge within the government space has been the continued use of mainframes that ran decades-old applications, some of which had been written in customized languages that fit the agency’s mission. One such federal agency that faced this challenge was The Department of Housing and Urban Development (HUD).
The Technology Modernization Fund (TMF), authorized in 2017 and disbursed by the General Services Administration (GSA), provided HUD with the opportunity and resources to modernize critical applications that ran on a mainframe.
Understanding the Need
To receive funds from the TMF, HUD needed a feasible solution for modernizing its legacy applications. These applications, which ran on the same Unisys mainframe, included:
- The Computerized Home Underwriting Management System (CHUMS)– Supports Home Ownership Centers (HOCs) staff in processing single-family mortgage insurance applications. This system conducts thousands of transactions with external resources daily.
- Credit Alert Verification Reporting System (CAIVRS)– Used by federal agencies to determine if a loan applicant has outstanding federal debt in default or foreclosure status.
- The Line of Credit Control System (LOCCS)– A disbursement and cash management system that distributes over $28 billion annually for grants and subsidies.
GovCIO Solution
Team GovCIO worked extensively to establish a holistic mainframe migration process for our federal customers that improves organizational performance while minimizing customer risk.
After identifying HUD’s need, GovCIO (formerly Salient CRGT) partnered with TSRI to proactively develop a working prototype for HUD that demonstrated the efficacy and viability of our process. HUD used this prototype to support its application for TMF funding. Once granted the funding, HUD awarded a sole-source contract to Team GovCIO.
Using our nine-step process, GovCIO rapidly migrated HUD from their COBOL-based mainframes to a modern Azure-based cloud service that can evolve with them, all while preventing negative impacts to their mission, day-to-day operations, and security.
- Define Project Goals
- Inventory System Components
- Identify Unique Technical Challenges
- Finalize Project Approach and Plan
- Create Test Strategy
- Migrate Data Stores
- Execute Automated Code Conversion
- Refactor Legacy Code
- Build Externals
Hear From Our Team
Results
The project was completed in two phases. Phase one comprised the Housing systems, CHUMS and CAIVRS. Phase-two covered the CFO system, LOCCS.
While our team planned the phase-one go-live for a three-day holiday weekend to allow the most time for any unforeseen challenges, we completed the system and data migration in just 14 hours. The go-live decision was made mid-day on the second day and the system went live on schedule with a normal production workload. The phase two migration was planned for a normal two-day weekend and was successfully completed in under 24 hours.
On day one of the system going live, CHUMS supported 25,356 users and 299,715 transactions with only three user problems reported. In the first 30 days after the LOCCS migration from the Unisys to Azure, the system disbursed just over $2.7 billion in program funds without error.
The successful completion of the automation modernization project solved one of the longest standing and most complex IT challenges facing HUD – moving these mission critical systems off the legacy mainframe platform.
As HUD was among the first to make use of the TMF funding for this type of initiative, HUD’s success can also be viewed as a significant endorsement of what agencies can achieve with TMF funding. GovCIO proved that a rapid, highly accurate migration of mainframe systems to Cloud services is possible, even for critical complex projects such as HUD’s.