[Beowulf] Is there really a need for Exascale?

Christopher Samuel samuel at unimelb.edu.au
Mon Dec 3 18:34:20 PST 2012


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 29/11/12 02:53, Prentice Bisbal wrote:

> I don't think this heretical. I think it's a perfectly legitimate 
> question we should be asking ourselves. Such a discussion can open
> a big can of worms on this list, and should probably be it's own 
> thread.

I will argue that there can be useful research work at exascale (which
is what I think the question is here).

For instance, the heart model Cardioid (which was a Gordon Bell
finalist this year) was able to model a single human heart at 0.1mm
resolution in *almost* real time on all of Sequoia with about 58%
efficiency (11+TF).

So if you wanted to simulate the effect of something over the period
of days, weeks or months you'll need to wait days, weeks or months.
Scaling that up means you could do a lot more testing in-silico.

> I've  got too much work to do today, so I can't weigh in as much as
> I should, but I will change the subject of this reply to start a
> new thread on this topic.

To really make a new thread you also need to get rid of the
References: and In-reply-to: headers as well, otherwise it just shows
up in the same thread in any sensible MUA.. :-)

cheers!
Chris
- -- 
 Christopher Samuel        Senior Systems Administrator
 VLSCI - Victorian Life Sciences Computation Initiative
 Email: samuel at unimelb.edu.au Phone: +61 (0)3 903 55545
 http://www.vlsci.org.au/      http://twitter.com/vlsci

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with undefined - http://www.enigmail.net/

iEYEARECAAYFAlC9YSwACgkQO2KABBYQAh8vFwCZAd27oAFSTOeEI4lztqPsMyFR
L48An2y+mxmixxEEVsvQyAbTQzxHx/js
=Q7kR
-----END PGP SIGNATURE-----


More information about the Beowulf mailing list