Hello Bugslist!
I'm an SA helping one of our WinBUGS proponents look into performance issues
with various hardware configurations. We used a single script,
characteristic of the type of operating we use on 4 machines.
The first machine is a dual processor 2.4Ghz Xeon DP server, with hyper
threading. The three other machines are desktops dialing in at a P3 700, a
P4 1.7Ghz and P4 2.4Ghz.
Here are some basic numbers:
Processor Operating Sys. Processing time
2.4Ghz XeonDP(2) Windows 2000 62 seconds
2.4Ghz Pentium 4 Windows XP 50 seconds*
2.4Ghz Pentium 4 Windows 2003 62 seconds
2.4Ghz Pentium 4 Windows 2000 62 seconds
2.4Ghz Pentium 4 Windows NT 4 62 seconds
1.7Ghz Pentium 4 Windows XP 50 seconds*
1.7Ghz Pentium 4 Windows 2000 62 seconds
1.7Ghz Pentium 4 Windows NT 4 62 seconds
700mhz Pentium 3 Windows XP 71 seconds
We found some interesting things.
The first is the use of the processor.
On a 700mhz machines, our WinBUGS 1.4 script was able to utilize 100% of the
CPU.
On the 2.4Ghz Xeon, 100% of hyper threaded CPU could be utilized, which in
actuality is not an entire CPU.
Once jumping up to 1.7, the utilization of the processor drops toward the
50% mark, and a bit less with the 2.4Ghz P4.
It would also seem by these numbers, that there is something about Winbugs
and XP that is faster than the same hardware running other versions of
Windows.
The other interesting aspect was that with the 2.4Ghz Xeon DP Hyper
threading processors, we could run anywhere from 1 to 8 simultaneous
instances of our script on that box without a performance hit. Each would
complete within the 62 second period.
Has anyone experienced anything similar?
Thanks in advance
Wilson Northrup
732.594.3255
------------------------------------------------------------------------------
Notice: This e-mail message, together with any attachments, contains information of Merck & Co., Inc. (Whitehouse Station, New Jersey, USA) that may be confidential, proprietary copyrighted and/or legally privileged, and is intended solely for the use of the individual or entity named on this message. If you are not the intended recipient, and have received this message in error, please immediately return this by e-mail and then delete it.
=============================================================================
-------------------------------------------------------------------
This list is for discussion of modelling issues and the BUGS software.
For help with crashes and error messages, first mail [log in to unmask]
To mail the BUGS list, mail to [log in to unmask]
Before mailing, please check the archive at www.jiscmail.ac.uk/lists/bugs.html
Please do not mail attachments to the list.
To leave the BUGS list, send LEAVE BUGS to [log in to unmask]
If this fails, mail [log in to unmask], NOT the whole list
|