I did add the CPU data from both system and user perspective. I think benchmarks would be required to get the difference between dedicated OSA and Virtual Switch - and a much higher traffic so differences are more obvious. "http://velocitysoftware.com/vsw.html";
>Date: Sun, 23 Oct 2005 11:47:03 +0200 >From: Rob van der Heij <[EMAIL PROTECTED]> > >On 10/23/05, Barton Robinson <[EMAIL PROTECTED]> wrote: > >> Is this kind of case study analysis useful? > >I would certainly think so. This allows you to go back and see who was >causing the traffic, etc. >What does bother me is that the reference is only to the vswitch >controller of the switch, not the switch itself. You could have more >than one vswitch controlled by the same vswitch controller (as in your >example) and Linux guests can have virtual NICs coupled to multiple >vswitches. You also want to know which is connected to what. >Do the monitor records tell how much resources are spent by CP to run >that vswitch so that we can obtain a kind of utilization level? > >Rob > >-- >Rob van der Heij rvdheij @ gmail.com "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 /************************************************************/
