Private partition specifications

Deeplearn

The deeplearn partition has been bought for use by Computing and Information Systems research staff and students

Node name	Nodes	Cores/node	Memory/node (MB)	Processor	GPU Type	GPU Memory	Slurm node type
spartan-gpgpu[086-088]	3	24	160000	Intel(R) Xeon(R) Silver 4214 CPU @ 2.20GHz	v100sxm2	32GB GPU RAM per GPU	dlg3
spartan-gpgpu[091-096]	6	32	234000	Intel(R) Xeon(R) Gold 6326 CPU @ 2.90GHz	A100	40GB GPU RAM per GPU	dlg4
spartan-gpgpu[065-071,132-143,160,166-169]	24	32	1000000	Intel(R) Xeon(R) Gold 6326 CPU @ 2.90GHz	A100	80GB GPU RAM per GPU	dlg5
spartan-gpgpu170	1	64	975000	Intel(R) Xeon(R) Platinum 8462Y+	H100	80GB GPU RAM per GPU	dlg6

Note

The deeplearn partition is available to specific Computing and Information Systems projects only. You can request access to it at the time of creating a Spartan account, or by submitting a help ticket.

To access the deeplearn partition:

#SBATCH --partition=deeplearn
#SBATCH --qos=gpgpudeeplearn

To specify that your job should only run on a specific node type, add a constraint to your submission script. e.g. to specify the V100SXM2 nodes, add #SBATCH --constraint=dlg3 to your submit script.

feit-gpu-a100

The feit-gpu-a100 partition has been bought for use by FEIT research staff and students.

Node name	Nodes	Cores/node	Memory/node (MB)	Processor	GPU Type	GPU Memory	Slurm node type
spartan-gpgpu[144-159,161-165]	21	32	1000000	Intel(R) Xeon(R) Gold 6326 CPU @ 2.90GHz	A100	80GB GPU RAM per GPU

Note

The feit-gpu-a100 partition is available to FEIT projects only. You can request access to it at the time of creating a Spartan account, or by submitting a help ticket.

To access the feit-gpu-a100 partition:

#SBATCH --partition=feit-gpu-a100
#SBATCH --qos=feit

fos-gpu-l40s

The fos-gpu-l40s partition has been bought for use by Faculty of Science research staff and students.

Node name	Nodes	Cores/node	Memory/node (MB)	Processor	GPU Type	GPU Memory	Slurm node type
spartan-gpgpu[181-192]	12	64	950000	INTEL(R) XEON(R) PLATINUM 8562Y+	L40S	48GB GPU RAM per GPU

Note

The fos-gpu-l40s partition is available to Faculty of Science projects only. You can request access to it at the time of creating a Spartan account, or by submitting a help ticket.

To access the fos-gpu-l40s partition:

#SBATCH --partition=fos-gpu-l40s
#SBATCH --qos=fos

fos

The fos partition has been bought for use by Faculty of Science research staff and students.

Node name	Nodes	Cores/node	Memory/node (MB)	Processor	Slurm node type
spartan-bm[182-185]	4	128	2001000	Intel(R) Xeon(R) Gold 6448H	avx512, sr

Note

The fos partition is available to Faculty of Science projects only. You can request access to it at the time of creating a Spartan account, or by submitting a help ticket.

To access the fos partition:

#SBATCH --partition=fos
#SBATCH --qos=fos