the p4 design is terrible - you can't compare cpus of different
architectures by clock speed alone. a p3 1.4 ghz isn't that far behind a p4
2.6 ghz in most real world situations, and a dual p3 1.4 would beat it
easily if you have lots of threads.
jason
Christopher L Merrill wrote:
Let's say I have an application that is fairly processor-intensive - which
means when it is running, it's using at least 50-75% of the available CPU.
It is also very thread intensive - running minimum of 500 threads, up to
well over a thousand. BTW, it's a Java program...and would be running on a
current Linux kernel, such as RHES4.
I am curious if I should expect to get better performance from, for
instance:
1) a slower dual-processor machine, e.g. dual 1.4GHz PIII
2) a faster single-processor machine, e.g. 2.6GHz P4
I realize any answer would have about a zillion caveats, but is
there any generally accepted wisdom along these lines?
TIA,
C
--
TriLUG mailing list : http://www.trilug.org/mailman/listinfo/trilug
TriLUG Organizational FAQ : http://trilug.org/faq/
TriLUG Member Services FAQ : http://members.trilug.org/services_faq/
