I totally concur with the recommendations to establish response time SLAs. Believe me, the business people do not view availability as simply "it's up but slow." They measure availability (rightly so) as the ability to get their jobs done.
SLAs must be agreed with the business users, and reviewed and updated periodically. Specific business activities will vary in their response time needs. Note that this is not mainframe-exclusive: it's for every aspect of IT. It just so happens mainframes are really good at setting SLAs and managing resources to them, with discriminating abilities. (In this situation discrimination is a very good thing.) In addition to the other excellent advice, while you're at it I would also look at the new LPAR group capacities you can set in z/OS 1.8 (and above) on the System z9. It's something to include as a factor when you crunch the z9 numbers. We live in a cheap hardware world, believe it or not, so very often getting nostalgic for a particular machine isn't economically the best idea. YMMV. I also concur that running at 100% for sustained periods isn't necessarily a bad thing if you're meeting your SLAs. But it sounds like you're missing the unofficial SLA -- you're getting complaints. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] ---------------------------------------------------------------------- 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
