Dear group,
I had approved for some of you to send calculations to the GPGPU queue when
the clustre was idle and did not need GPGPU jobs. Now, we need to clear the
cluster of jobs, because Roberto needs to run some things.
I think we need to ask the people at FAS to have a "backfill" low priority
job queue that immediately is KILLED whenever a gpgpu job is sent. I think
that would work. If one wants to run a non-GPGPU job, one should be able, as
to fill the cluster, but having no guarantees of completion. What do you
think?
In the meantime, Sigi or whoever is running, please terminate your GPGPU
jobs, as Roberto and company need to use the cluster for graphics cards. (I
know I told you to backfill, but it is not working as it is interrupting the
GPU research).
Thank you,
Alan
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