David Day: Very interesting thread. Since I read IBM-MAIN in DIGEST mode, I missed it yesterday during real-time.
I don't think I have ever heard of anyone considering slowing down response time to make it as slow as a mainframe for benchmarking purposes. Neat idea. I have only used it many times to prove problems in mainframe program code that are unable to handle devices (both DASD and TAPE) that are so fast due to incorrect logic in the mainframe program. The mainframe Linux inability a few years ago to boot on FLEX-ES with that problem is a famous example. Just a reminder that you can benchmark and see how many cache hits you are getting from the FLEX-ES controller cache by using the "display ckdcachestats nnn" command and "clear ckdcachestats nnn" command in the FLEX-ES CLI. You may want to bring more specific discussions on topics like that over to the FLEX-ES listserv. (Thanks Alan Altmark for pointing out that FSI has had ESCON attachment capability for quite a while now. Oh, and a FLEXCUB can be more than just a DASD control unit for a conventional mainframe. It can be tape, 3172s, and unit-record gear, too. Concurrently.) -- Gary Eheman Fundamental Software, Inc. http://www.funsoft.com ---------------------------------------------------------------------- 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
