Thanks Guillermo, I didn't spot those options.
But I'm still confused as to what they really mean.
Can you split an L-L test into parallel threads over say 8 cpu's, and get
the answer ~ 8 times faster ?
I had thought that the L-L algorithm was strictly sequential.
I guess you can exploit OpenMP and get a 2x or 3x/4x speedup in the short
pieces of code where stuff is really happening in classical "parallel", but
my guess would have been that would end up getting you about a 1.2X - 1.5X
speedup overall. I apologize I'm making these wild generalisations with no
knowledge of the L-L code.

To answer Max: yes I'm sure for throughput I should just run 8 copies of
single-thread L-L, and I hope to be able to try that. But I have a funny
situation here: almost exclusive access to some 4p/8p/16p Itanium2 machines
for periods of up to say two weeks, plus access to anything from 32p to 256p
slices of bigger machines for say a few days. So an MP version of Glucas L-L
that truly scales would let me finish full L-L runs inside these time limits
on all the machines.

Cheers,
Greg E 



-----Original Message-----
From: Guillermo Ballester Valor [mailto:[EMAIL PROTECTED] 
Sent: Saturday, 2 April 2005 7:51 PM
To: [email protected]
Cc: Greg Edwards
Subject: Re: [Prime] Multi-processor LL tests

Er..

Of course, the link is not correct, I forgot to change 'localhost' by 
'www.oxixares.com':

http://www.oxixares.com/glucas/doc/Configure-Make-and-Install.html#Configure
%20Make%20and%20Install

El S�bado, 2 de Abril de 2005 10:24, Guillermo Ballester Valor escribi�:
> On page:
>
>
http://localhost/glucas/doc/Configure-Make-and-Install.html#Configure%20Mak
>e%20and%20Install
>
> you will see the multithread options you can use at './configure' time.
You
> can use Sunmp (for Sun compilers), Openmp (for compilers supporting that)
> and POSIX threads.
>

-- 
Guillermo Ballester Valor
[EMAIL PROTECTED]
Ogijares, Granada  SPAIN
Linux user #117181. See http://counter.li.org/
Public GPG KEY http://www.oxixares.com/~gbv/pubgpg.html

 
_______________________________________________
Prime mailing list
[email protected]
http://hogranch.com/mailman/listinfo/prime

Reply via email to