GPUs on the HPC

This page covers the Graphics Processing Unit (GPU) resources available on the cluster.

See also

GPU Hardware details for different types of GPUs available on Discovery.

The gpu partition is the general GPU resource for HPC users looking to use a GPU; multigpu is the alternative, where more than one GPU are accessible.

See also

Learn more about partitions.

Anyone with a cluster account has access to the gpu partition. However, you must submit a ServiceNow ticket requesting temporary access to multigpu provided sufficient need and preparation.

Note

When working with shared computational resources, it is important to remember not to leave the jobs idle.

Name	Requires Approval?	Time in Hours (Default/Max)	Submitted Jobs	GPU per Job Limit	User Limit (No. GPUs)
gpu	No	4/8	4/100	1	4
multigpu	Yes	4/24	8/100	8	8

Important

Consider the compatibility of the GPU, as some programs do not work on the older k40m or k80 GPUs.

Execute the following command to display the non-Kepler GPUs that are available:

sinfo -p gpu –Format=nodes,cpus,memory,features,statecompact,nodelist,gres

This indicates the state (idle or not) of gpu-types and could be helpful to find one that is idle. However, the command does not give real-time information of the state and should be used carefully.