Benchmarking the Shift to Mainframe Agility
Not long ago, enterprises were still allowing their mainframe Application Development and Delivery (AD&D) teams to languish in slow, antiquated waterfall development. Some pundits even opined that it was reasonable to focus their Agile and DevOps efforts only on customer-facing systems of engagement—while actually disinvesting in their critical back-end systems of record.
But enterprise CIOs have quickly and dramatically woken up. They now recognize that mainframe agility is a fundamental requirement for business agility, because the vast majority of their projects involve mainframe code and data—and the speed with which you can respond to the market’s new digital value demands will always be constrained by your slowest AD&D team.
So major initiatives are underway everywhere to make mainframe AD&D teams as responsive to the business as every other platform’s team. Under these initiatives, IT organizations are embracing new processes and tools that speed mainframe development, enable Agile and DevOps best practices, and empower a new generation of digital artisans to work with complex and often under-documented COBOL codebases.
As they pursue these initiatives, however, IT leaders quickly find themselves asking two big questions: 1) “How am I doing?” and 2) “Can I do better?”
Well, we have some answers.
Metrics for Mainframe Success
Compuware’s approach is simple—and reflects the data-driven approach being embraced across virtually every category of human endeavor these days.
Outcomes often correlate with behaviors. Sometimes these correlations are easy to spot. Sometimes they aren’t. But if you want to know why an outcome occurs, it makes sense to look at the behaviors that led up to it.
And that’s what we’re doing with our zAdviser offering. We’re capturing behaviors across the mainframe software development and delivery lifecycle, and then using machine learning to correlate those behaviors to key objectives in the shift to mainframe agility—specifically improvements in quality, velocity and efficiency.
We’re already discovering lots of actionable correlations. Sometimes a correlation involves team members under-utilizing a particular feature of a particular tool. Sometimes it involves anomalous bugs related to the way two particularly complex applications interact.
Whatever the underlying causes happen to be in any given situation, the key is to have the right data and the right models to make sense of them. And that’s what we’re delivering with Compuware zAdviser.
We’re also aggregating anonymized data from across our customer base to establish benchmarks for the move to mainframe agility. This provides you with a useful perspective on how you’re performing in relation to peers facing similar challenges. Data aggregation also helps us zero in on behaviors and process changes that have the greatest impact on mainframe modernization efforts.
Unifying the Cross-platform Digital Value Chain
Of course, mainstreaming the mainframe entails more than just removing the esoteric differences of the platform and making it nimbler and more accessible to next-gen developers. Ultimately, you want your preferred DevOps tool chain for your distributed and cloud platforms to extend its benefits to the mainframe—because new digital value often depends on modifications to application logic running on multiple inter-dependent tiers.
That’s why, rather than offering a single monolithic solution, Compuware works closely with popular best-in-class solution providers like Atlassian, Amazon, Jenkins, Parasoft, SonarSource and XebiaLabs. We’ve also just announced a new partnership with CollabNet VersionOne that will help large enterprises incorporate their mainframe development activities into a unified Value Stream Management approach across all platforms.
We’re doing this by integrating ISPW, our DevOps-enabled source code management (SCM) solution for the mainframe, with CollabNet VersionOne’s Continuum—which tracks, automates and orchestrates software value streams across the DevOps lifecycle. The initial result of this partnership will be full cross-platform visibility into how your development work is progressing to delivery.
Ultimately, however, partnerships like these will generate even richer, broader data for our machine learning algorithms, giving you ever-greater insight into how you’re doing and how you can do better.
Want to Join Us?
For mainframe agility to change, that culture has to change as well. We all have a lot to gain as a community by learning from each other. And that’s happening more and more. We’re seeing a much more energetic mainframe community emerge on social media, and we’ve felt the vibe at live events around the world.
That’s why we’re inviting you to become part of the zAdviser community. The more data we have, the more quickly and confidently we can pinpoint opportunities to continuously improve our mainframe practices.
So if you’re not engaged with zAdviser yet, please give us a call. We all need metrics. And we all have metrics to share. Together, we can successfully advance the cause of mainframe agility—and all come out winners.
This post originally appeared on LinkedIn.