One of the better things about the shift to cloud computing is that it creates an opportunity to address the sins of the past. One of those major transgressions has been the divide between proponents of mainframe and distributed computing models, which in their zeal to be right all too often came close to tearing the IT house asunder.
But now we’re starting to see an interesting opportunity to start unifying the management of mainframe and distributed computing systems across the enterprise. Compuware this week is trying to start IT organizations down that path with the release of new application performance management (APM) offerings. The first new offering consists of Outage Analyzer, a service that identifies the causes of any outage from any particular cloud service that an IT organization is depending on. The second offering is an update to Compuware APM for Mainframe, a set of APM tools for the mainframe that now includes the PurePath technology for tracking application components at an atomic level that Compuware gained when it acquired dynaTrace in 2011.
According to Steve Tack, vice president of product management for Compuware’s APM business, integrating the dynaTrace technology that was originally developed for distributed systems running in the cloud with mainframe APM tools is a significant step toward creating a unified APM framework. That’s critical because instead of managing mainframes and distributed systems in isolation, the longer-term trend is going to be to unify the management of applications that are increasingly comprised of workloads running on different classes of platforms. While that approach results in better optimal use of IT infrastructure, holistically managing all those workloads remains a major IT challenge.
Unfortunately, Tack notes that many of these borderless workloads are only coming together for the first time when they hit the end user’s browser, which makes it difficult for IT to determine what particular service may be responsible for compromising an end-user experience. The simple fact is that Web applications in particular are becoming borderless entities, with different components running on distributed computing systems, mainframes and now third-party data centers in the cloud.
The biggest inhibitor to managing all this right now, however, may not be the technology, but rather the animosities that have built up over the years between proponents of mainframes and distributed computing systems. Fortunately, it looks like the rise of more complex applications is about to finally force the reconciliation issue.