<html><head><style type="text/css"><!-- DIV {margin:0px;} --></style></head><body><div style="font-family:times new roman, new york, times, serif;font-size:12pt"><div style="font-family: times new roman,new york,times,serif; font-size: 12pt;">I did read an article about re-compiling for the target cpu. It had to do with hyper-threading cpu Pentium 4 HT. You have to test the multi-tasked. <br><br>For example: <br>A few years back we had processes that took too long , so I did an experiment. (This was at a college) There was a process that took an hour ans 20min to process 2000 students. So we broke up the data set and ran it in parallel I started both processes about 5 seconds a part. it took 29min to process 2000 students. Quite an improvement. <br><br>So there are a few thing that the 64bit dual core offerers:<br><br>1. Data Transfer <br>2. Large Calculations. <br>3. Multi Tasking <br><br>In my test the operating system switched tasks
between both tasks, that was HP/UX with 5 cpus. In this case there is just 2.<br><br>The other issue is handling memory. 64bit can point to more memory than 32bit. <br><br><br><div style="font-family: times new roman,new york,times,serif; font-size: 12pt;">----- Original Message ----<br>From: Aaron Kulkis <akulkis3@hotpop.com><br>To: "mdlug@mdlug.org" <mdlug@mdlug.org><br>Sent: Tuesday, October 30, 2007 3:09:34 PM<br>Subject: [mdlug] [Fwd: Re: [opensuse] 64 bit vrs 32 bit advantages speed etc.]<br><br>
<br><br>-------- Original Message --------<br>Subject: Re: [opensuse] 64 bit vrs 32 bit advantages speed etc.<br>Date: Tue, 30 Oct 2007 14:42:23 -0400<br>From: Buddy Coffey <<a ymailto="mailto:bcoffey@XXXXX.XXXX" href="mailto:bcoffey@XXXXX.XXXX">bcoffey@XXXXX.XXXX</a>><br>To: <a ymailto="mailto:opensuse@opensuse.org" href="mailto:opensuse@opensuse.org">opensuse@opensuse.org</a><br><br><br>>>> Has anyone run benchmarks on an identical system with 32 bit vrs 64
bit?<br>>>><br>>>> <br>Here is a benchmark of 32 bit vs 64 bit openSuSE 10.3 on a Core 2 Duo<br>system.<br><br>I have both 32 bit and 64 bit versions installed in separate partitions<br>on the same system with all updates applied.<br><br>I ran my computationally intensive code GEMACS (<a href="http://www.gemacs.com" target="_blank">http://www.gemacs.com</a>)<br>compiled natively for both 32 and 64 bits with the Intel 9.1 compilers.<br>The code performs little I/O and zero graphics. The code is<br>multi-threaded and fairly well balanced, hence, the total CPU numbers<br>are about 2X the elapsed time numbers.<br><br>64 bit:<br> Elapsed time: 59.316 sec<br> Total CPU time: 112.43 sec<br><br>32 bit:<br> Elapsed time: 59.749 sec<br> Total CPU time: 105.67 sec<br><br><br>I would be glad to run other comparisons if anyone would like them.<br><br>Buddy Coffey<br>Applied
Research Associates<br>Computational and Applied Electromagnetics<br><br><br>_______________________________________________<br>mdlug mailing list<br><a ymailto="mailto:mdlug@mdlug.org" href="mailto:mdlug@mdlug.org">mdlug@mdlug.org</a><br><a href="http://mdlug.org/mailman/listinfo/mdlug" target="_blank">http://mdlug.org/mailman/listinfo/mdlug</a><br></div><br></div></div><br>__________________________________________________<br>Do You Yahoo!?<br>Tired of spam? Yahoo! Mail has the best spam protection around <br>http://mail.yahoo.com </body></html>