GPUs on the HPC¶

See also

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

The NVIDIA GPUs available on gpu-equipped partitions.¶
GPU Type	GPU Architecture	Memory (GB)	Tensor Cores	CUDA Cores	Public Nodes (x # GPUs)	Private Nodes (x # GPUs)
P100	Pascal	12	N/A	3,584	12(x3-4)	3(x4)
V100 PCle	Volta	32	640	5,120	4(x2)	1(x2), 16GB
V100 SXM2	Volta	32	640	5,120	24(x4)	10(x4), 16GB 8(x4), 32GB
T4	Turing	15	320	2,560	2(x3-4)	1(x4)
A100	Ampere	41 & 82	432	6,912	3(x4)	15(x2-8)
Quadro RTX 8000	Turing	46	576	4,608	0	2(x3)
A30	Ampere	24	224	3,804	0	1(x3)
RTX A5000	Ampere	24	256	8,192	0	6(x8)
RTX A6000	Ampere	49	336	10,752	0	3(x8)

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.

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.