On 20 Oct 2001, at 9:44, Mikus Grinbergs wrote:

> Interesting to compare the performance numbers given for the Itanium
> running Glucas v2.8c against my Thunderbird running mprime v21.4 :
> 
>  -  At the smallest FFT length, the Itanium is WAY faster.

Probably a "large cache" effect.
> 
>       this performance difference decreases until
> 
>  -  At FFTs 640K-2048K, the Itanium is a little bit faster

Yes, by 640K FFT run length you need > 2MB cache to run with no 
significant access to "real" memory.
> 
>       then the performance difference increases until
> 
>  -  At the largest FFT length, the Itanium is noticeably faster
> 
> 
> Is it memory-bandwidth that lets the Itanium pull ahead at the
> large FFT lengths ?
> 
I would suspect that the main cause is that the Thunderbird has a 
fairly limited TLB - as the amount of memory involved in the 
workspace increases, the proportion of recursive TLB lookups will 
increase faster than it does with Itanium and so the memory 
access will effectively slow down.

Regards
Brian Beesley
_________________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ      -- http://www.tasam.com/~lrwiman/FAQ-mers

Reply via email to