Kees, I suspect the loopers don't give WLM enough time to react to the WLM Resource Group cap. If I understand your below the 2 MSU's is just under 1% of the total. With a more normal workload, less consistent arrival rate, you may get more desirable results. I guess unless you're running loadrunner or some such generator. I have the RG max. for your RG for this machine and based on your below at 846.
________________________________ From: "Vernooij, CP - SPLXM" <[email protected]> To: [email protected] Sent: Wednesday, August 29, 2012 10:49 AM Subject: Re: WLM Resource Group not working. 1) yes, and mainview confirms this. 2) Velocities. 3) The LPAR involved runs on a 2817-503, so a 3 CP, 204 MSU z196, of which I want to give and limit 2 MSU to CICS Devl/Acc. 2000000 SU/hr = 555 SU/sec. I also tried 3 CP percents (option 3) which also equals 2 MSU (204/3*.03=2) but this gives similar results. WLM is not willing to cut the users in the RG to the desired consumption. Yes, that is what I understand: the specified cap is the limit for all users in the RG. Kees. "Staller, Allan" <[email protected]> wrote in message news:<[email protected]>... > 1) have you placed the tasks in the appropriate service classes to be linked with the RG? > 2) Are you using transaction goals or velocity goals. > 3) Perhaps you might want a processor cap as opposed to a SYSPLEX cap (Option 3). Are you saying that your sysplex has an aggregate of 2 million SU/sec? How many LPARs, CECs are involved. > > The last time I tested, I found that the RG cap was shared between all users of the RG, regardless of the service class involved > > Al Staller | Z Systems Programmer | KBM Group | (Tel) 972 664-3565 | [email protected] > > > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Vernooij, CP - SPLXM > Sent: Wednesday, August 29, 2012 8:51 AM > To: [email protected] > Subject: WLM Resource Group not working. > > Hello list, > > > > I implemented a Resource Group to limit the CPU consumption of our CICS Devl and Acc systems, but I can't get the RG to do what I want. > > > > I want to limit the group to 2 MSU, so I gave the RG a max of > 2000000/3600=555 SU/sec (Option 1, sysplex wide) and started a CPU looper (iefbr15) in the SC connected to the RG. In Mainview I see that WLM is capping the looper, but it is still allowed to consume 2000 SU/sec. If I start a second looper, the RG starts consuming even more CPU. > > > > Why does this not work as I expected and how do I limit the group to 555 SU/sec? > > > > Thanks for your help, > > Kees. Koninklijke Luchtvaart Maatschappij NV (KLM), its subsidiaries and/or its employees shall not be liable for the incorrect or incomplete transmission of this e-mail or any attachments, nor responsible for any delay in receipt. Koninklijke Luchtvaart Maatschappij N.V. (also known as KLM Royal Dutch Airlines) is registered in Amstelveen, The Netherlands, with registered number 33014286 ******************************************************** ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
