[H-GEN] Hyperthreading and CPU utilisation
Michael Anthon
michael at anthon.net
Mon Aug 25 03:20:19 EDT 2003
[ Humbug *General* list - semi-serious discussions about Humbug and ]
[ Unix-related topics. Posts from non-subscribed addresses will vanish. ]
> 23.9 + 76.1 adds up to 100 in my book.
>
> > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
> > 25073 root 18 0 328 328 272 R 99.9 0.0 0:33.50 cat
>
> And that's reasonable if you're using cat to generate load, which is
> limited by its read() and write() calls.
>
Yeah, I understood this and wasn't really questioning it.
I guess the problem I'm trying to get my head around is that from my
(admittedly limited) understanding of HT, 2 processes running on the one
HT CPU will not run as fast as 2 processes running on 2 separate CPUs
(assuming that the process, or to phrase it another way, if a single
process running on an HT processor can utilise more then 50% of the
available processing time of the execution engine then running a second
similar process on the same CPU will obviously not double the
performance since it will have to throttle back the processing of the
first thread.
I am using cacti to monitor CPU usage on my servers, using snmp to read
the jiffy counters from /proc/stat. My first surprise was that on this
system these counters return 4 times the expected values although with
hind sight it makes more sense. The most confusing thing though is that
a single process running flat out as with the cat mentioned earlier then
it ends up being by both the jiffy counter and top method (since both
derive from the /proc/stat values) running at 25% of full load. I just
get the feeling that it's not really 25%, probably really closer to 50%.
This is the issue I can't reconcile since there doesn't seem to be a way
to measure it.
Cheers
Michael
--
* This is list (humbug) general handled by majordomo at lists.humbug.org.au .
* Postings to this list are only accepted from subscribed addresses of
* lists 'general' or 'general-post'. See http://www.humbug.org.au/
More information about the General
mailing list