[H-GEN] Hyperthreading and CPU utilisation

[Michael Anthon]

[ Humbug *General* list - semi-serious discussions about Humbug and     ]
[ Unix-related topics. Posts from non-subscribed addresses will vanish. ]

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.  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.

The system is an Oracle DB server and is going from a Sun E250 with 2 x 
400MHz UltraSparcs, 640M of RAM and a single 40M scsi bus driving disks 
and tape to a DL380 G3, with dual 2.8G Xeons, 2.5G of RAM and 3 
different disk controllers (2 x raid array controllers and a scsi 
controller for the tape).

At a rough guesstimate we are increasing the CPU grunt about 6-7 times 
(hard to compare) and the bus speed about 8 times plus splitting the 
system across 3 different buses.  Current system takes 14 hours to do 
the daily backup on a *good* day!  I'm quite looking forward to this 8^)

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/