WHOAA.
Dear group,
Look at these LARGE MEMORY processors. This might be super useful for
certain group applications (or lazy programmers)
Alán Aspuru-Guzik | Associate 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 |
http://about.me/aspuru
---------- Forwarded message ----------
From: David Turner <dpturner(a)lbl.gov>
Date: Thu, Jun 16, 2011 at 3:21 PM
Subject: [mpp-users] New extra-large-memory resource available
To: mpp-users(a)nersc.gov
Dear NERSC User,
NERSC has recently installed and tested a unique resource
for users who require more memory than is currently provided
by our other computational systems. We have deployed a single
node in the Carver/Magellan complex with the following
characteristics:
* 4 8-core Intel Xeon X7550 (Nehalem EX) 2.0GHz processors
* 1TB DDR3 memory
* Same software environment as Carver login nodes
Access to this 32-core node is via the "mag_xlmem" batch queue on
Carver. Note that this queue allows batch jobs to share the node,
rather than have exclusive access. Jobs must have the following
characteristics to use this queue:
* No more than 8 cores (ppn=8)
* Memory size *must* be specified; no less than 4GB per core,
and no more than 800GB per job
* No more than 48 hours walltime
Users are limited to 2 jobs running in this queue.
The above limits will probably be adjusted depending on user
feedback.
For more information on submitting jobs to this new resource,
please see:
https://www.nersc.gov/users/computational-systems/carver/running-jobs/memor…
Carver batch queue limits and policies are documented here:
https://www.nersc.gov/users/computational-systems/carver/running-jobs/queue…
--
Best regards,
David Turner
User Services Group email: dpturner(a)lbl.gov
NERSC Division phone: (510) 486-4027
Lawrence Berkeley Lab fax: (510) 486-4316
_______________________________________________
mpp-users mailing list
mpp-users(a)nersc.gov