On Thu, 31 Mar 2005 09:57:40 -0500, Calvin Ward <[EMAIL PROTECTED]> wrote: > Just to be clear, are you saying it is better to have more simultaneous > requests in that scenario, or less instances, or both?
Both. Although load testing real applications on real configurations is really the only way to find out what is optimum. My experiences with Solaris and the SPARC architecture would lead me to suggest no more than one instance per 2 CPUs (not based on any empirical evidence, but based on writing compilers for the Sun 4 chipset back before it was called SPARC). I don't know enough about Windows multi-threading and the Intel processor to make a recommendation for that platform. I would be surprised if one instance per processor (with 3-5 simultaneous requests each) turned out to be more efficient that one instance per 2 processors (with 6-10 sim reqs per instance) - on any processor architecture. Process switching is likely going to be more costly than thread switching. -- Sean A Corfield -- http://corfield.org/ Team Fusebox -- http://fusebox.org/ Got Gmail? -- I have 50, yes 50, invites to give away! "If you're not annoying somebody, you're not really alive." -- Margaret Atwood ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~| Find out how CFTicket can increase your company's customer support efficiency by 100% http://www.houseoffusion.com/banners/view.cfm?bannerid=49 Message: http://www.houseoffusion.com/lists.cfm/link=i:4:201037 Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4 Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4 Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4 Donations & Support: http://www.houseoffusion.com/tiny.cfm/54