https://community.hpe.com/t5/operating-system-hp-ux/capacity-planning-for-web-server/m-p/2596515#M854625 Capacity planning for web servers is rather difficult as the data is bursty and follows a power tail distribution. Does anyone have any statistical formulas for assessing the amount of data that must be analyzed versus the resultant error margins in the performance or capacity data that you generate from an analytical queueing model?<BR />Mott Given, mgiven@dsdc.dla.mil Wed, 17 Oct 2001 13:54:21 GMT Mott Given 2001-10-17T13:54:21Z https://community.hpe.com/t5/operating-system-hp-ux/capacity-planning-for-web-server/m-p/2596515#M854625 Capacity planning for web servers is rather difficult as the data is bursty and follows a power tail distribution. Does anyone have any statistical formulas for assessing the amount of data that must be analyzed versus the resultant error margins in the performance or capacity data that you generate from an analytical queueing model?<BR />Mott Given, mgiven@dsdc.dla.mil Wed, 17 Oct 2001 13:54:21 GMT https://community.hpe.com/t5/operating-system-hp-ux/capacity-planning-for-web-server/m-p/2596515#M854625 Mott Given 2001-10-17T13:54:21Z https://community.hpe.com/t5/operating-system-hp-ux/capacity-planning-for-web-server/m-p/2596516#M854626 WOW, that's a lot of big words. Just kidding. Building a web server is more complicated than building a server for apps like oracle. What I've done, is build big fast boxes, get everything running, and tweak the crap out of it. I'd love to have modeling software to do it, and I know that there are supposed tools out there to do it (with supposed very underlined). Wed, 17 Oct 2001 14:02:18 GMT https://community.hpe.com/t5/operating-system-hp-ux/capacity-planning-for-web-server/m-p/2596516#M854626 harry d brown jr 2001-10-17T14:02:18Z