SQL Server 2005 performance testing with hyperthreading and MAX DOPhttp://rentacoder.com/CS/blogs/real_life_it/archive/2006/04/18/SQL_Server_2005_performance_testing_with_hyperthreading_and_MAX_DOP.aspxThere's been alot of negative publicity about hyperthreading (see "Hyperthreading hurts server performance, say developers" which is all over the net...including: http://news.zdnet.co.uk/0,39020330,39237341,00.htm) We had been experiencing disapointingen-USCommunityServer 2.0 (Build: 60217.2664)re: SQL Server 2005 performance testing with hyperthreading and MAX DOPhttp://rentacoder.com/CS/blogs/real_life_it/archive/2006/04/18/SQL_Server_2005_performance_testing_with_hyperthreading_and_MAX_DOP.aspx#467Thu, 27 Apr 2006 03:09:33 GMT2b679553-572d-4457-89da-485055e17287:467arbertThe biggest negative impact of hyperthreading is mainly related to the operating system. &nbsp;In the days of W2k, Microsoft would sometimes recommend you disable hyperthready because the base operating system saw the logical CPUs as physical CPUs and would allocate work accordingly--this could cause things to slow down quite a bit. <br> <br>You'll find that most operating systems support it quite well now. <br> <br>As far as MAXDOP, it really depends on the query--sometimes it will make a difference, sometimes it won't. &nbsp;Another thing that really affects the performance of a parallel plan is the other activity that is happening on the SQL Server at the same time. &nbsp;The estimated execution plan can be quite different from the actual execution plan.