How Will Federal IT Overcome Challenges with Legacy Systems?
Overview: A 2018 Executive Briefing from Federal News Radio that was sponsored by Compuware outlines challenges federal IT executives face with legacy systems in a digital age and what they can do to overcome those challenges.
Government agencies face a plethora of challenges related to their legacy systems. But how you define “legacy” determines your strategy for overcoming those challenges.
“The challenge with legacy systems is not that they don’t work. Rather, if anything, they work too well and reliably,” Tom Temin, anchor and columnist, writes in a new Executive Briefing for Federal News Radio sponsored by Compuware.
Mainframe systems in particular work so well and reliably that the code they run could be considered digital gold. COBOL powers entire organizations and is their foundation for digital transformation with customer-facing technologies of engagement.
For instance, in the Executive Briefing, Temin points out that agencies like the IRS still rely on the power of mainframe systems, some of which were programmed in the ‘60s. The reason for this, and for why many large enterprises—primarily financial and insurance firms—also still rely on the mainframe, is strategic not negligent.
Mainframes remain the most powerful systems of record available to large enterprises undergoing transformations in the digital age. With unmatched reliability, availability and security, there’s a reason these backend monoliths still run most of the business-critical applications and data that enterprises are built on.
In fact, research from Forrester commissioned by Compuware found 72 percent of mainframe-powered firms’ customer-facing applications are completely or very reliant on mainframe processing. And the same research estimates an increase from 57 to 64 percent of these firms will be running more than half their mission-critical workloads on the mainframe in 2019.
Despite this, pressure to modernize is making it difficult for agencies to continue investing in the maintenance of legacy code and systems like the mainframe, Temin writes. New mandates and incentives through the Modernizing Government Technology Act, for instance, are pressuring agencies to scale back datacenters, go cloud-first and reduce the cost of maintaining legacy systems to free up money for innovation.
Other factors for why modernization is being pushed include a growing skills shortage as experts who maintain these systems retire as well as difficulties in extracting legacy applications from databases that exist in highly specific formats to be used “in digital services which exist as new, interactive web applications,” Temin writes.
Clearly, there are challenges that need solving when it comes to legacy systems. But is a “rip, rewrite and replace” strategy really the best option for mission-critical applications and data running on your mainframe?
Temin asked several federal IT executives what they thought about recruiting new talent for legacy systems, whether Agile and DevOps are possible on the mainframe, and how legacy systems of record can be leveraged in the digital age.
Read the briefing to learn what federal IT executives are thinking, from ensuring modernization initiatives are strategized around business needs, not a trend of legacy-systems replacement, to concerns of how agencies can continue to leverage systems like the mainframe while battling reduced budgets and attrition.
“Sanjay Gupta, chief technology officer of the Small Business Administration, said, ‘There is inherent value’ in the legacy code. ‘The challenge is, how do you continue the sustainability of that?’” Temin writes.
Fortunately, there is plenty of hope for federal IT executives wondering what they should do to balance modernization with legacy systems, where working code is gold.
“‘DevOps toolchains and agile processes can gain dominion over the mainframe,’” Temin writes quoting Compuware CEO Chris O’Malley, “letting organizations add to the digital virtues of the reliability, performance, security, and transactional efficiency of modern mainframes like the IBM