Current thoughts from IBM ATS which can probably be considered canon on this
http://www-03.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/WP101229 My thoughts for what they are worth. You may not see a large amount of benefit now with your configuration on a z10 but it is probably the right direction to go so yes enable it after doing appropriate planning. It defaults to being enabled at release 1.13. If you get it enabled now without any problems then when you do go to hardware that would allow you to benefit as you go forward with confidence and get that out of the box on the new hardware. I enabled HiperDispatch day one on our z9 to z10 implementations (smallest machine for us was a -709) and never looked back still on when we went to z196. It has been a "good thing". I would suggest to be reasonably current on service at a recommended recent quarterly RSU. Ask IBM through the formal support channels directly where you can get service recommendation specific to your environment before making changes. Be prepared to monitor and measure the results of that change. We recently consolidated even more LPARs onto our z196 and found the following HiperDispatch APAR's were important (at least to us). We initially had some difficulty with one workload breaching an SLA after consolidation. We engaged with the smart folks in IBM ATS right here in Gaithersburg and they steered us to these one of which we had already installed but the other two were added on top of RSU1111. OA36459 in particular seemed to be important for us to be able to get full capacity utilization out of the box. * OA35989 - On a large CEC with low utilization, except for a small test partition running with HD=YES, vertical low processors may not be unparked, even though there is sufficient demand on the small partition and there is a large amount of free capacity on the CEC - Routine which calculates free capacity suffered an overflow due to large amount of unused capacity * OA35860 - Running with HD=YES, vertical low processors may be unparked even though there is no unused capacity available on the CEC - WLM calculations of available capacity did not account for capacity used by *PHYSCAL partition * Impact is only when there is high Physical LPAR management time * OA36459 - Closed 12/1/2011 - Not calculating the capacity used by vertical mediums and vertical low processors correctly Best Regards, Sam Knutson, GEICO System z Team Leader mailto:[email protected] (office) 301.986.3574 (cell) 301.996.1318 "Think big, act bold, start simple, grow fast..." -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of gsg Sent: Friday, February 24, 2012 1:03 PM To: [email protected] Subject: Hiper Dispatching We have a 2097-504(E12). Does anyone know if turing on Hiper Dispatching on a single book box machine would make a difference in performance? ==================== This email/fax message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution of this email/fax is prohibited. If you are not the intended recipient, please destroy all paper and electronic copies of the original message. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
