A New Generation of Application Performance Management

“It is all a matter of perspective!”  This old adage rings particularly true when trying to solve a complex IT problem that spans mainframe, distributed, and mobile platforms.

Mainframe programs are increasingly interfacing with distributed systems to support consumer facing web and mobile applications. When a transaction is initiated via a mobile phone, the path it takes from the edge of the Internet and deep into the mainframe and back is highly complex — and ripe with possibilities for bottlenecks along the way.  Should a problem occur in any one of the platforms along the application delivery chain, teams typically lack true visibility into the offending transaction. An unfamiliar platform can quickly become a “black box,” void of any window into how an application is impacting its performance.  Without visibility into the problem, hard-to-answer questions ultimately lead to unproductive discussions and finger pointing.

A little Kumbaya, please

The scenario above highlights the need to understand multiple perspectives. The teams’ understanding of “performance overhead,” for example, is based upon their own perspectives. From a mainframe point of view, performance overhead is primarily concerned with CPU utilization. To that end, the mainframe performance team would want insight into how the distributed platforms are driving mainframe workload.  And from a distributed perspective, it is about end-user response time, so the distributed performance team would want visibility into the mainframe that provided their transaction processing and its impact on end-user response time.

Both the mainframe and distributed performance teams would benefit from a new generation of application performance management (APM), one that provides end-to-end transaction visibility, from the edge of the Internet, through distributed systems, and its backend processing by CICS and DB2.  A common view of the performance problem allows the distributed and mainframe IT teams to work together more effectively to solve issues.

Embrace the “New Normal” of Mainframe

The end-lesson here is that IT teams need to embrace multiple perspectives from the various platforms their applications depend upon.   Application complexity is increasing as mainframe programs are being interconnected with distributed systems.  Consumer facing applications are driving additional mainframe workload.  And this is only the beginning. This is the New Normal of Mainframe and it’s here to stay.

The following two tabs change content below.

Spencer Hallman

Spencer Hallman is Product Manager for Compuware's Strobe and iStrobe performance tools. Previously, he was a Subject Matter Expert for Mainframe Performance and Field Technical Support for Strobe. His diverse experience over the years has also included programming on multiple platforms, providing technical support and working in the Operations Research field. He has a Master of Business Administration from Temple University and a Bachelor of Science in Computer Science from the University of Vermont.