[Beowulf] Re: building a new cluster
schuang21 at yahoo.com
Thu Sep 2 11:51:46 PDT 2004
--- Joachim Worringen <joachim at ccrl-nece.de> wrote:
> SC Huang wrote:
> > 5. I only have wall-clocking timing of my code on various
> platforms. I
> > don't know how sensitive it is to cache size. I guess the bigger
> > the better, because the code is operating large arrays all the
> If your arrays are really large, than it's the memory bandwidth that
> counts, not the cache size. Or your ability to implement
> (blocked) versions of your code (probably with the assistance of
> libraries like ATLAS).
The arrays are large, but I have made "continous memory access" as much
as I can. So cache reuse should be ok. I once tested this using a Cray
compiler. The "cache miss" ratio seemed reasonable.
> Just a side note on this: we have a 32-node dual-Athlon cluster with
> Myrinet in the lab as well as an SX-6i (single processor SX-6). We've
> seen our financial modelling codes run faster on the SX-6i than on
> cluster (with a non-trivially parallized version). And in contrast to
> the cluster, the SX-6i is running non-stop for months (as long as we
> have it here). It's much smaller and more energy-efficient, too - but
> it's not cheaper.
Thank you for the comments. Is SX-6 a shared memory machine? May I ask
how many processors were used when it ran faster then the cluster? Just
> Joachim Worringen - NEC C&C research lab St.Augustin
> fon +49-2241-9252.20 - fax .99 - http://www.ccrl-nece.de
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit
Do you Yahoo!?
Yahoo! Mail - 50x more storage than other providers!
More information about the Beowulf