[H-GEN] Hyperthreading and CPU utilisation

Anthony Towns aj at azure.humbug.org.au
Mon Aug 25 12:33:21 EDT 2003


On Mon, Aug 25, 2003 at 11:18:46PM +1000, Michael Anthon wrote:
> Anthony Towns wrote:
> >You can measure performance you're getting, but you can't really measure
> >the unused capacity remaining; this is because how much you have available
> >depends on what exactly you want to do with it.
> This is pretty much the conclusion I came to, it just took me a while to 
> wrap my thinking around it.  It's hard, having become so used to being 
> able to accurately measure performance.  

Except you can't: cache interactions and sizes, memory and disk sizes
and speeds, and interrupts will all affect the speed of a single program;
and if you've got multiple programs running, refreshing caches, swapping,
processor and disk scheduling algorithms and prioritisation will all
have non-trivial impact on performance.

Hyperthreading really isn't much harder than all that, although it
certainly is new and different.

> I should have a better idea 
> next week when we bring this thing online in a production system 
> (cutting over from old server to the new one this weekend all going 
> well) and I can really hammer it with proper data.

Do give us a report :)

Cheers,
aj

-- 
Anthony Towns <aj at humbug.org.au> <http://azure.humbug.org.au/~aj/>
I don't speak for anyone save myself. GPG signed mail preferred.

       ``Is this some kind of psych test?
                      Am I getting paid for this?''
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 307 bytes
Desc: not available
URL: <http://lists.humbug.org.au/pipermail/general/attachments/20030826/d57bd2cd/attachment.sig>


More information about the General mailing list