>From: "Kinnear, Mike" <[EMAIL PROTECTED]> > >Has anyone running Linux under VM attempted to compare or evaluate the >differences between the Velocity Software's ESALPS and IBM's VMRTM/VMPRF? >What were your conclusions as far as function and value? > >Thank you for your help.
In reference to replys to this post, I would not have compared ESALPS to FCON, I would have compared zMON to FCON (or RTM). (See "http://www.velocitysoftware.com/zmon.html";). zMON is meant for the VM entry level user that does not need full function performance monitoring and capacity planning - and the price is EXTREMELY HARD TO BEAT.... (See published world wide prices at: "http://www.velocitysoftware.com/price.html";) ESALPS is a more functional product for capacity planning and accounting. I've looked for anyone actually measuring inside Linux using the FCON/toolkit solution and can NOT find anyone willing to show real data, and no such REAL data has been presented at any conference (share, tech conf, CMG). As many of the new linux installations have found - something that "works" for one or two images is quite different than something that "works well" on many images. I find it curious that IBM's emphasis on FCON/toolkit is for something that nobody seems able to use to measure inside Linux, and Linux REALLY IS the future for zVM. I think a strategy check is in order. There are many presentations showing actual data from ESALPS given at many different conferences, (and of course see the recent Performance REDBOOK SG24-6926) there should be no doubt about if ESALPS works or not. Working with installations with 30-40 linux images - all being measured by ESALPS, with low overhead, I have no problem showing it works. See Velocity Software's "presentation page" at "HTTP://velocitysoftware.com/present.html"; for some of the presentations showing real data. One more point for the "non-entry" level installations: The CPU numbers coming from inside Linux are inflated for any measurement products running on Linux under VM - or even often in LPARs. One of my standard presentations shows the inflation at 10 fold. It is VERY VERY hard to perform useful capacity planning or accounting based on wrong CPU data. ESALPS corrects the numbers and produces valid data. Does any other vendor publically recognize this problem, much less fix it? Many people still seem to be completely unaware that these numbers are wrong. And I keep thinking this is old news. I would welcome a valid comparison - but mostly, I would like to know if anybody has ever tested the FCON/toolkit/RMFPM method for measuring many servers from inside Linux and can share the resource requirements in public.... "If you can't measure it, I'm Just NOT interested!"(tm) /************************************************************/ Barton Robinson - CBW Internet: [EMAIL PROTECTED] Velocity Software, Inc Mailing Address: 196-D Castro Street P.O. Box 390640 Mountain View, CA 94041 Mountain View, CA 94039-0640 VM Performance Hotline: 650-964-8867 Fax: 650-964-9012 Web Page: WWW.VELOCITY-SOFTWARE.COM /************************************************************/
