[sdpd] Multi-core and SDPD

[Armel Le Bail <alb...@cristal.org>]

Hi,

Calculations on PC taking advantage from parallelism will obtain speed
improvement in the next years (or even now) in spite of the limitation
to < 4GhZ for a single-processor. Visit :
http://www.theinquirer.net/?article=24188
http://www.x86-secret.com/popups/articleswindow.php?id=125
the latter in French...

For now, the Pentium XE is in fact identical to a dual-Xeon in
performance though 2 times more expensive...

Anyway, Pentium D820 sounds interesting (or some equivalent,
if not better, AMD).

Thinking of dual-CPU + dual-core seems even already accessible
(i.e. 4-core).

Just lacking are the software for SDPD able to use them
efficiently. Any Monte-Carlo-based software should benefit
of multi-core. At least, I have the project to rebuild parallel
versions for ESPOIR, McMaille and GRINSP before the end
of 2005. The latest version of the Intel Visual Fortran compiler
allows for that.

Other coming events about SDPD and multi-core technology ?

Are some of you already using multi-core PCs for applications
in crystallography specially adapted, and which ones ?

Armel



 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/sdpd/

<*> To unsubscribe from this group, send an email to:
    sdpd-unsubscribe...@yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/