[ale] Hyperthreading

[James Sumners]

I would but the box does more than just compile things.

On 15 Aug 2004 23:17:30 -0400
"James P. Kinney III" <jkinney at localnetsolutions.com> wrote:

> On Sun, 2004-08-15 at 22:51, James Sumners wrote:
> > That is very odd. I have a box with dual 2.0Ghz HT processors and it can
> > compile MySQL in four minutes and some change.
> 
> You have two physical processors. Some things do work OK with HT. Some
> aspects of program compiling should while others should thrash the
> cache. Test this by turning off the HT in the bios and recompile from a
> fresh MySQL tarball and time it. Then reboot an turn the HT back on and
> do it again and publish your results. 
> 
> Even better would be to add a third step of disabling one processor and
> running just HT and get that timing data.
> 
> And while you're at it, turn of HT and do it all a fourth time just for
> a baseline.
> 
> None of my dual CPU machines are even _close_ in capabilities so I can't
> test anything like this with my rigs. (How does one compare a dual
> Opteron 1.4 GHZ with a dual PIII 550 and a PIV 2.2GHz w/HT?)


-- 
James Sumners
http://ug.dyndns.org/~jsumners/

"All governments suffer a recurring problem: Power attracts pathological
personalities. It is not that power corrupts but that it is magnetic to the
corruptible. Such people have a tendency to become drunk on violence, a
condition to which they are quickly addicted."

Missionaria Protectiva, Text QIV (decto)
CH:D 59