>I'm sure all capacity planning people must have problems selling this to their >people.
For 25 years, in my case. I still get push-back from management. Especially, these days with the flaws in LSPR: 1. They introduced new workloads for the z/990 (WebSphere for example). And, they never calibrated them with the older models. So, how do do when know that WebSphere is xx times what it would be on an older model? 2. They didn't evaluate all configurations. I'm not sure where they stopped on a z9 (20 or 24 CP's), but on a z/990 they stopped at 16. Everything above the stopping point was straight-lined. They used to admit that on the LSPR web page, but I don't think they do anymore. But, the increase in capacity as an engine is added is NOT linear! The degradation as each engine is added is. So, if you re-examine the data in that light, you can get a difference of 3000+ MIPS on a z/990 at the high end. Which figures are correct? I don't know, but IBM's are highly suspect. Fortunately, I work at a shop that doesn't need much more than a 311-312. But, what about shops that need bigger boxes? They are making capacity, performance, AND software decisions on suspect data! And, nobody knows the true cost of LPAR management at that end of the spectrum, either! We just know that it goes up based on logical CP's and LPARS! I'm going to get stuck with a real mess when we get large enough! I don't really blame IBM for the whole problem. Configurations are getting too big to (cost) effectively evaluate them. Customers are asking for even larger (look how quickly IBM had to release the S54 -- almost a year earlier than planned). But, they should be more open (and up-front) with us as to what they evaluated. Not bury it; not mask it with one sentence that has since disappeared! This is going to haunt us as we move forward! And, the platform already has acceptance issues! - -teD 300,000 Kilometres per Second Not only is it a good idea! It's the LAW!!! ---------------------------------------------------------------------- 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
