Date: Sat, 28 Jun 2003 05:42:26 +0200 From: Marco Wertejuk <wertejuk@mwcis.com> To: current@freebsd.org Subject: Re: Hyperthreading Message-ID: <20030628034226.GA3321@maeko> In-Reply-To: <20030627181836.GA81335@node1.cluster.srrc.usda.gov> References: <20030627181836.GA81335@node1.cluster.srrc.usda.gov>
next in thread | previous in thread | raw e-mail | index | archive | help
In order to give some values to compare: I was recently running an P4 3.06 GHz with Hyperthreading (actually it was Linux, but I will repeat the tests with current when I have time for such games :) Running ubench 0.32 on this HTT enabled machine (i865PE) showed some interesting details when accessing memory: ubench (without parameter): cpu: 111742, memory: 178587, average: 145164 ubench -s (single cpu): cpu: 101993, memory: 135693, average: 118843 With Hyperthreading disabled: ubench (without parameter): cpu: 100878, memory: 133539, average: 117208 ubench -s (single cpu): cpu: 102041, memory: 135296, average: 118668 As you can see, the system has an amazingly improved memory performance when hyperthreading enabled (about 25%) but the overall calculating speed increases only about 9%. Probably you can run ubench on your HTT system earlier than me and post the results here. -- Mit freundlichen Gruessen, Marco Wertejuk - mwcis.com Consulting & Internet Solutions
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030628034226.GA3321>