Hi !
Long time no see....
In the past days - with the great help htcondor-admin@ , thanks ! - I've managed to set up a perfectly working Condor pool with docker universe working and I'm able to run GPU jobs from inside a Docker container, everything is OK !
The next step would be to enable complete virtual machines, not only containers, but so far I haven't found a working way of enabling GPU access from inside a virtual machine.
I've read rumors about "PCI passthrough" solutions that makes this desired setup working but found no consistent description on that.
Does anybody runs a Condor pool with virtual machines (instantiated on job request) and with jobs inside those able to access GPUs ?
Thanks a lot in advance for any hint,
Gergely
This e-mail and any files transmitted with it contain confidential and may contain privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and delete this e-mail. Any unauthorized use, copying, disclosure or distribution of the material in this e-mail is strictly forbidden. |