Problem solve Get help with specific problems with your technologies, process and projects.

How to reduce the engine utilization of a certain CICS region

I am currently working with a customer whose application is constrained to a single CICS region TS1.3. This CICS is now at 70% of the MIPS available on a single engine but we have another engine available to this MVS.

I am considering setting SUBTSK=1 as this is a very heavy VSAM system but cannot find out exactly how this will help. I understand that the VSAM writes, opens and some CI/CA processing will execute under a different TCB but am not sure what address space this TCB will execute in or how I can monitor its performance from Omegamon or SMF.

In your experience does this option significantly help reduce the engine utilization of the CICS giving us a bit more headroom.

For a CICS region running a workload which makes heavy use of VSAM (non-RLS) update operations VSAM subtasking can improve throughput by off-loading some of the VSAM processing to another TCB. This TCB runs in the same CICS region and is managed by CICS: it is known as the CO (for concurrent)TCB. In an optimal case performance measurements have demonstrated a single CICS region driving 1.2 processors. This does not translate to a 120% increase in throughput because there is significant instruction overhead switching the work from the main CICS TCB to the CO TCB, and back again. Each TCB switch costs somewhere between 1K and 2K instructions, depending on how busy the TCBs are.

You can see how much work is being off-loaded to the CO TCB by looking at the CICS statistics and monitoring data for the dispatcher.

Dig Deeper on IBM system z and mainframe systems

Have a question for an expert?

Please add a title for your question

Get answers from a TechTarget expert on whatever's puzzling you.

You will be able to add details on the next page.

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.