FERMI box.
Group, the CDI people have worked hard to give us access to an nVidia FERMI
box, it would be interesting to benchmark it, e.g. with GPU and MP2 codes,
it probably files!
A.
Alán Aspuru-Guzik | Assistant Professor
Harvard University | Department of Chemistry and Chemical Biology
12 Oxford Street, Room M113 | Cambridge, MA 02138
(617)-384-8188 |
http://aspuru.chem.harvard.edu
---------- Forwarded message ----------
From: Hanspeter Pfister <pfister(a)seas.harvard.edu>
Date: Thu, Mar 18, 2010 at 3:18 PM
Subject: Access to pre-release Fermi board
To: Hanspeter Pfister <pfister(a)seas.harvard.edu>
Cc: "Robert E. Parrott" <parrott(a)seas.harvard.edu>du>, Seppo Seppo <
seppo(a)seas.harvard.edu>
As you may know, we have a Fermi engineering sample from NVIDIA. This is not
the final version of the hardware or drivers, and performance will be much
slower than the final product. Any results published need to be cleared with
NVIDIA and should include this disclaimer. (It's best to actually not
publish performance results from this board at all.)
But the board allows us to get our codes running and to try out some of the
new architecture features. We're still ironing out kinks, but we hope to
have it up and running by the end of the week. The machine has Linux and
Windows installed.
Please send email to ircshelp(a)seas.harvard.edu if you would like to have
access to the Fermi. By doing so you agree with the conditions outlined
above.
Thanks,
- Hanspeter