What’s Your IBM i Application Debt?
March 31, 2014 Alex Woodie
Enterprise IT tool vendor Micro Focus recently did a survey that found mainframe shops around the world had an average of $11 million in application debt, and that the backlog of unfinished and un-started IT projects had increased nearly 30 percent in the last two years. While the survey didn’t look into the IT debt situation at IBM i shops, the numbers can’t be far behind.
The idea of application debt is an interesting one. The analyst firm Gartner first coined the phrase a few years ago to refer to the amount of resources it would take for an organization to complete the backlog of application modernization projects necessary to get the organization’s IT systems up to parity with a hypothetical median state for a given organization in a given country and a given industry. It’s a bit of an academic exercise, to be sure. But broaching the subject in this way–on the back of the almighty dollar–can also be a useful lens to view the problems associated with application modernization, or the lack thereof.
The study–which involved 590 CIOs and IT managers in nine countries and which Micro Focus paid the research firm Vanson Bourne to conduct–doesn’t provide any definitive answer for mainframe-related application debt problems. Spending more money, clearly, has to be part of the equation, but that alone won’t do it. Instead, the research points to the need for a multi-pronged approach to modernization that encompasses tools, technology, processes, and skills, says Derek Britton, director of product marketing at Micro Focus.
“Some of these applications were designed and built 40 years ago for the way businesses operated those days, so perhaps the process and technology choices and certainly many of the skills are not aligned for the way business needs to work today,” Britton tells IT Jungle. “Tackling the way appropriate tooling and appropriate skilling came out as key issues. People don’t think they have all the right skills in order to get the work done.”
A modern application today could mean several things. It may have a mobile or Web interface, or run in the cloud. Applications with monolithic codebases are looked down upon by those who ascribe to the model-view-controller (MVC) hierarchy. A modern mainframe app may have one or more components that run within a .NET or a JVM environment, and ideally these are maintained alongside the COBOL and CICS, regardless of which mainframe subsystem they touch. “It’s possible to achieve these with technology from Micro Focus and others,” Britton says. “But it isn’t always readily evident to the people who are stuck maintaining these mainframe applications that that’s even possible.”
In the mainframe world, COBOL is king, and the reliance on that language is directly analogous to the predominance of RPG in the IBM i world. Many of the same problems afflicting mainframe shops–including the prevalence of decades-old procedural code maintained in a green-screen environment and the lack of younger people with mainframe administrative skills–are also big hurdles to modernization in IBM i shops.
There’s one thing that the $11 million debt figure for mainframe shops most definitely does not signify, Britton says, and that’s a growing willingness to ditch the mainframe and the COBOL apps in favor of some “modern” system–say SAP running on Intel iron, today’s enterprise IT status quo. Mainframes are considered core parts of an organization’s fabric and a key competitive differentiator, just as they are in IBM i shops, and that’s not changing any time soon.
“If you’re just swapping out and replacing with a new system, what you’re typically not doing is replacing it with something better. It’s just equivalent, which of course your competitors can do at the same time,” he says. “There’s no way that any replacement system was going to do as good a job and it was going to take years to replace and then it wasn’t going to move the needle at all and it was actually going to cost a lot of money.”
As a tools developer, it’s no surprise that Micro Focus bends the discussion toward how tools can help. To that end, the company is focused heavily on equipping COBOL developers with a modern Eclipse-based environment to streamline coding, debugging, and deployment alongside “modern” languages like Java and C++.
Implementing agile processes can play a big part in cutting application debt, whether the organization runs on mainframes, IBM i servers, or (as we hear so often today) iPads. “Their processes were perfectly adequate for systems that were developed and delivered 20 to 30 years ago, but I’m not sure IT can get away with as small a number of deliveries per year as they used to,” Britton says of mainframe shops. “Maybe they have to go agile and think about doing more instantaneous deliveries. Maybe the mainframe isn’t geared for that the way the processes are currently structured. But somebody who can short circuit those process–without sacrificing quality, of course–is going to help improve time to delivery and time to market.”
So, how does the $11 million in average mainframe modernization translate into the IBM i world? While both systems speak EBCDIC, there’s no direct connection between these Big Blue second cousins. For starters, mainframe shops tend to be a bit bigger. While the IBM i server is not strictly used by small and midsized businesses (as it is often misperceived to be), the prevalence of the IBM i server in SMBs would definitely skew the numbers downward. Without being privy to the specifics that went into the equation–such as the number of lines of code requiring modernizing or the number of programmers who need retraining–it would be a meaningless guess. With that said, what harm would a meaningless, off-the-cuff estimate pose? Let’s say it’s half the mainframe debt, or $5.5 million.
Numbers aside, the important takeaway for IBM i shops here is to begin the conversation around application modernization, to help organizations think about the problem in terms of technology, tools, skills, and process, and to figure out a way to break the problem down into bite-sized pieces. “It’s just a matter of understanding what’s possible,” Britton says. “As soon as they understand the possibilities, the modernization journey is much more readily accessible than perhaps they realized.”