Hi, All, Say a cicsA address space is assigned a service class STCP1. From SMF 99 data I can see that when transaction volume is high enough, cicsA would be be moved into server mode.
When in server mode, cicsA is moved into an internal service class like $SRMSxxx. cicsA is managed by WLM according to transaction service class goal. I have some questions. 1. Does WLM take cicsA into account when calculating STCP1 PI? In my opinion, it shouldn't. cicsA is now managed according to transaction service class goal. It makes no sense to include it. 2. Does STCP1 PI in RMF report at cicsA server mode time range include CICSP1? I think it shouldn't. The same reason. 3. Does resource usage of STCP1 in RMF report include cicsA? A little tricky here. I think it should. Because from my experience APPL% of STCP1 in RMF report does include the cpu usage of transactions. BTW, I find that all the transaction management and internal service classes of WLM is complicated and its logic not well documented. And I'm doubting whether it is really better than velocity goals. I'm considering managing CICS using velocity service classes only at least in our test systems. Best Regards, Johnny Luo ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
