https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Correctly-Measuring-HT-Performance/m-p/973008#M5522 <DIV></DIV> <DIV></DIV> <DIV>Since we've been seeing a number of HT questions lately, I thought I would point out this article about measuring performance on Hyper-Threading Technology enabled systems.</DIV> <DIV></DIV> <DIV><A href="http://www.intel.com/cd/ids/developer/asmo-na/eng/technologies/threading/66456.htm" target="_blank">http://www.intel.com/cd/ids/developer/asmo-na/eng/technologies/threading/66456.htm</A></DIV> <DIV></DIV> <DIV>I don't know anything about queueing theory, so I can't comment on the theoretical aspects of the article. However, I do know thatthe advice to measure the right thing is always relevant.</DIV> <DIV></DIV> <DIV>-- clay</DIV> Wed, 28 Jan 2004 00:05:16 GMT ClayB 2004-01-28T00:05:16Z https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Correctly-Measuring-HT-Performance/m-p/973008#M5522 <DIV></DIV> <DIV></DIV> <DIV>Since we've been seeing a number of HT questions lately, I thought I would point out this article about measuring performance on Hyper-Threading Technology enabled systems.</DIV> <DIV></DIV> <DIV><A href="http://www.intel.com/cd/ids/developer/asmo-na/eng/technologies/threading/66456.htm" target="_blank">http://www.intel.com/cd/ids/developer/asmo-na/eng/technologies/threading/66456.htm</A></DIV> <DIV></DIV> <DIV>I don't know anything about queueing theory, so I can't comment on the theoretical aspects of the article. However, I do know thatthe advice to measure the right thing is always relevant.</DIV> <DIV></DIV> <DIV>-- clay</DIV> Wed, 28 Jan 2004 00:05:16 GMT https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Correctly-Measuring-HT-Performance/m-p/973008#M5522 ClayB 2004-01-28T00:05:16Z