Things do get interesting even on z9 processors with more than 24 processors. IRD does not prevent N-way overhead and not far above 24 at least for some typical commercial workloads that curve can take a wicked bend upwards. When captured time starts climbing for no reason consider that more processors not only provide less MIPS they increase the cost of existing workloads which is visible time in CPU per transaction TCB/SRB. When you get out to 16 - 18 CP's and expecting growth in a single image it is time to start working on further data sharing exploitation or other means to divide that workload.
zPCR is definitely worth getting some in house expertise on and consulting when considering CEC configurations and what you can get out of them. zPCR is FREE after all it just takes some time to learn and make sense of what you are looking at. Do the education! http://www-03.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/PRS1381 IBM has some folks who do this for a living who can help. The folks at WSC are pretty sharp cookies. The cool thing about z/OS and Sysplex is that you can grow horizontally because the cost to keep growing vertically in a single partition gets a little pricey even with today's technology. The closest thing I have seen in terms of real working technology to z/OS Sysplex with data sharing from a competitor is Oracle RAC. On System z a new processor generation with faster engines would not be a bad thing both for speed and to get another break on software MSUs. Hey I know what I want for Christmas:-) Best Regards, Sam Knutson, GEICO System z Performance and Availability Management mailto:[EMAIL PROTECTED] (office) 301.986.3574 "Think big, act bold, start simple, grow fast..." -----Original Message----- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Ted MacNEIL Sent: Sunday, July 15, 2007 6:20 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Re: Links to decent 'why the mainframe thrives' article >A z9 UP is closer to 600 MIPS. The latest claim is 585. But, that doesn't matter, since IBM has only benchmarketted to 24 CP's. The rest has to be done with zPCR. Which is suspect, since they changed the LSPR workloads (bits and types) and have not explored beyond. - Too busy driving to stop for gas! ==================== 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: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
