The strategic decisions public-sector CIOs must make over the next six to 18 months are being shaped by three realities:
Why the Mainframe?
1. Applications must be developed more quickly. In an increasingly digital world, rapid delivery of new applications is a must. Legislatures are mandating it. Executives are staking their political careers on it. And constituents require it. IT organizations must become more agile, which means they have to rapidly capture requirements and ensure the performance of code in production. 2. Money remains tight. IT budgets aren’t keeping pace with escalating demands for new mobile, analytic and social capabilities. CIOs must therefore figure out how to do even more with even less—despite the fact that they’ve already probably wrung as much efficiency as they can from the virtualization of their distributed infrastructure. 3. IT personnel are undergoing a generational shift. A growing number of IT employees are both “digital natives” and mission-minded. The skills and passion of these new employees are definitely advantageous when it comes to meeting demand for new digital services. CIOs, however, must be proactive about managing a new workplace culture and maximizing employee retention in a competitive skills marketplace. It’s fashionable to assume that these realities are best addressed by embracing the cloud and/or trying to eke more value out of on-premise commodity infrastructure. But CIOs who examine the facts will find that they can achieve better results with a strategy that receives far less hype—but delivers far greater value. That strategy is to more aggressively leverage the mainframe. There are several reasons why the mainframe is a compelling platform for public-sector IT’s new workloads. First, government IT organizations already have significant investments in the mainframe. It makes sense economically and logistically to leverage these existing investments in application code, data and operational best practices in the pursuit of new digital objectives. Second, the mainframe is a technically matchless platform. Its performance, scalability, reliability and security are far beyond that of any distributed or cloud infrastructure. In fact, despite the incalculable investments made in these commodity platforms, they have never come close to delivering what the mainframe does. That’s why the most critical workloads in both the public and private sectors continue to run almost exclusively on mainframes. Third, despite popular misconceptions to the contrary, the mainframe’s economics are superior to other platforms. The main reason for this misconception is that mainframe costs are fully visible, while total ownership costs for distributed platforms are largely hidden. But you can literally double the workloads running on a mainframe without adding a single operations staffer. That’s not the case with commodity infrastructure. So the incremental cost of workload growth on the mainframe is relatively low—especially if you manage your MSUs (million service units) well. The cloud can be even worse, because there are no economies of scale. Every increment of utilized capacity generates cost. That can lead to some very unpleasant surprises at the end of the month when utilization spikes unexpectedly. Finally, the mainframe can be highly agile. Yes, traditional processes that have accrued over the years tend to make application changes on the mainframe slower than in distributed or cloud environments. But that’s not an inherent aspect of the platform. It’s a result of habit. Those habits can be changed to streamline DevOps on the mainframe and make it as agile as any other platform—but with greater confidence and less cost. While there are significant benefits to be gained by leveraging the mainframe as a strategic platform for new workloads, those benefits may escape the faint of heart. That’s because participation in the mainframe renaissance requires:
Authentic leadership. As-a-service solutions are great for obtaining certain types of undifferentiated functionality on the cheap. But CIOs who simply “follow the crowd to the cloud” will pay what everybody else pays—and risk what everyone else risks. To capitalize on the mainframe opportunity, CIOs have to take time to understand that opportunity and communicate it to the team. Investment in new tools. Millennials don’t have years of COBOL analysis experience. And they will find “green screen” tools unsettling. They will require tools that empower them to work on the mainframe just as they would on Windows and Linux systems. Veteran mainframe pros can also benefit from tools that help with agile best practices—because, as it turns out, old dogs actually enjoy learning new tricks. IT de-siloing. At most agencies, the mainframe team is segregated from the rest of IT. This no longer makes sense, because applications are increasingly built across platforms. Ultimately software development teams should have the flexibility to host application components wherever it makes the most sense to do so—whether on the mainframe, in the data center or out in the public cloud. Vendor aggression. Most CIOs have become complacent when it comes to mainframe software—or have limited their negotiations to price and service concessions. But to take the mainframe to the next level, CIOs need their mainframe software vendors to be much more innovative and to support new IBM z Systems features such as Linux and Java processing. These are not high prices for the superior capabilities the mainframe offers. CIOs who pay them will be far better able to cope with escalating stakeholder demands even as agency IT budgets remain constrained. The result will be better use of taxpayer dollars to provide the most resilient and secure delivery of next-generation public digital services. Claire Bailey is the director of federal, state and local solutions for Compuware. She was previously the CIO of Arkansas for eight years.