On Wed, 19 Dec 2007 21:05:38 +1000, Shane <[EMAIL PROTECTED]> wrote: >I have a bit of empathy with both Dave and Ted. >Personally I like the idea of a percentage - even in an asymmetric >sysplex. Much easier. >Isolating specific work to a particular CEC ain't that hard. Setting >meaningful sysplex-wide goals gets interesting, but that's really a >different issue and has been covered before. >However, I have had requests along the lines of - "make sure X doesn't >get more than yy% of an engine" (or similar). This is easiest >accomplished with raw service units. Accommodates adding engines with >only minor errors - upgrades mean dragging out the slide rule again. >
It's nice to have options. :-) But I have to agree with Ted on this one, especially as it relates to this thread and guaranteeing a minimum level of service. What if you have a small footprint and someone comes along and decides you need a much bigger footprint to support some new application. After the upgrade you've now guaranteed your "unimportant" test workload many more MIPS (and I'm sure they will get used also). Mark -- Mark Zelden Sr. Software and Systems Architect - z/OS Team Lead Zurich North America / Farmers Insurance Group - ZFUS G-ITO mailto:[EMAIL PROTECTED] z/OS Systems Programming expert at http://expertanswercenter.techtarget.com/ Mark's MVS Utilities: http://home.flash.net/~mzelden/mvsutil.html ---------------------------------------------------------------------- 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
