[Beowulf] Benchmark reality check please
Robert G. Brown
rgb at phy.duke.edu
Wed Dec 22 19:55:36 PST 2004
On Wed, 22 Dec 2004 steve_heaton at ozemail.com.au wrote:
> That RGB fellow's BenchMaster suite would seem to give the CPU/RAM
> side of things a good workout. I'll give that a burl. Until recently
> LMBench seemed the go but BenchMaster seems to be a step up (more
> flexible)? (Someone *other* than RGB's opinion would be nice =P ) Maybe
> the Netpipe suit for all sorts of juicy network numbers?
Fine, be that way;-)
Actually, lmbench is without a doubt a superior benchmark from a
technical point of view. It's just a PITA to download, build, and run,
as you have to go through bitkeeper and use a networked SCCS
auto-extraction tool. Mine you can just grab and go -- if you're
running one of the distros I've built for you can probably even install
the rpm and just run it, if you aren't going to play the compiler tuning
optimization game.
lmbench, OTOH, gives you LOTS of very precise measurements on
EVERYTHING, and they've been tuning their timing harness MUCH longer
than I have (to be fair to Larry and Carl).
rgb
--
Robert G. Brown http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567 Fax: 919-660-2525 email:rgb at phy.duke.edu
More information about the Beowulf
mailing list