I am faced with a grave performance problem in one of my CICS regions, particularly with a few transactions (CICS/ESA 4.1 -- not sure when we'll go to TS). These transactions do a lot of I/O and the wait is at resource FCIOWAIT and JCIOBLOK (on system journal J01). To circumvent the journal wait problem, I am thinking of making the JNLSYNCWrite from YES to NO (so that the journalling will be asynchronous).
- Will this give any improvement?
- Are there any potential risks in doing this?
- I haven't coded the SYSWAIT=STARTIO in the JCT. Will this cause the tasks to wait for the buffers to be flushed (I read that STARTIO will be the default)?
- The primary and secondary journals switch in the interval of 2-3 minutes during peak time (worst case is 38 seconds).
- The IBM PTFs PN80128/UN89909 have been applied.
This type of query should be raised with the CICS Service organisation.
However, I am SERIOUSLY concerned that you seem, after the year end, to be planning to use a then out-of-service product: you WILL be seriously exposed by using a then unsupported version of CICS.
Dig Deeper on IBM system z and mainframe systems
Related Q&A from Robert Crawford
For better mainframe capacity planning, how do I convert CPU hours to MIPS? And is there a way to calculate the relationship between MIPS and MSUs? Continue Reading
I have two years of experience in mainframe technology, currently working as a mainframe developer. I want to change to Java technology. Continue Reading
I want to replicate DB2 from the mainframe to an AIX box since it's cheaper and the copy can be used for testing. Is this possible? Continue Reading
Have a question for an expert?
Please add a title for your question
Get answers from a TechTarget expert on whatever's puzzling you.