[Coral_current] GPU cluster (fwd)

[Tim Oates]

FYI, we should be using this resource.  If there are reasons why it is
difficult to use or does not meet your needs, let me know.

   - tim


---------------------------------------
Tim Oates, Professor
Department of CS and EE
University of Maryland Baltimore County
(410) 455-3082
https://coral-lab.umbc.edu/oates/

---------- Forwarded message ----------
Date: Sat, 2 Apr 2022 17:22:24 -0700
From: Hamed Pirsiavash <hpirsiav at umbc.edu>
To: Tim Oates <oates at cs.umbc.edu>
Subject: GPU cluster

Dear Tim,

Hope all is going well.
As you already know, UMBC has received MRI funding from NSF, and purchased a
relatively large GPU cluster called ?ada?. It is now fully operational and
freely available to the whole campus. It contains:
104 high-end GPU cards (NVIDIA 2080-Ti, RTX 6000, and RTX 8000)
26TB of fast SSD storage
1,248 CPU threads
5,760GB of memory

Please consider using this valuable resource that is freely available for
the whole campus.

You and your students may ask for an account through the following form.
Please select ?ada? for the ?System? field and then select ?Account or Group Request?:
https://doit.umbc.edu/request-tracker-rt/doit-research-computing/

You may learn more about ada and how to use it at the following link:
https://hpcf.umbc.edu/ada/environment/

Yours,
Hamed