(RADIATOR) counterintuitive benchmarks

Guðbjörn S. Hreinsson gsh at centrum.is
Mon Apr 19 10:06:33 CDT 2004


> however, surely a faster machine (dual xeon, hyper-threaded enabled)
> should be able to hit the server at a harder rate than the slower
machines.
> in fact the faster 2-xeon machine takes longer to stress radiator than
> three other separate slower machines.

I have had a chance before to compare single-threaded processes on
hyper-threaded xeon cpu's versus pentium III xeons. In all cases the PIII
performed better.

The same is not at all true for multithreaded applications. So I am not
really
surprised with your finding.

> surely the stopevers affact all these machines. further, the 2-xeon is not
> running other servers or heavy clienst, whereas the slower machines do
> occasionally service other tasks...!

Try it with MT disabled. And you are probably only using one of the cpu's.
-GSH

--
Archive at http://www.open.com.au/archives/radiator/
Announcements on radiator-announce at open.com.au
To unsubscribe, email 'majordomo at open.com.au' with
'unsubscribe radiator' in the body of the message.


More information about the radiator mailing list