Table of contents

1. Cluster configuration
2. CPU queue/partitions
3. GPU queue/partitions
4. Special nodes
5. Using SLURM commands

Cluster configuration

Number	Manufacturer	Cores	RAM	GPUs	GPU Memory	Name
20	Intel	32	768 GB			arctrdcn{001-020}
3	Intel	96	1.5 TB			arctrdhm{001-003}
20	AMD	64	512 GB	1xNvidia 2080	11GB	arctrdagn{001-020}
16	AMD	128	512 GB	2xNvidia A40	45GB	arctrdagn{031-046}
1	Nvidia DGX-1	256	1 TB	8xNvidia A100	40GB	arctrddgxa001
2	Nvidia DGX-1	192	1 TB	8xNvidia A100	80GB	arctrddgxa{002-003}
2	Dell	40	192 GB	4xNvidia V100	32GB	arctrdgn{001-002}

CPU queue/partitions

Partitions	Nodes	Time limit	Priority	CPUs	Memory	GPUs	Limitations	Preemption	Purpose
qTRD	arctrdcn001-arctrdcn016	5d 8h	N/A	32	768GB	N/A	N/A	N/A	General purpose (production) computing
qTRDGPU	arctrdagn[001-020,031-046]	5d 8h	N/A	64	512GB	gpu:RTX	N/A	N/A	General purpose (production) computing, occasional GPU computing
qTRDHM	arctrdhm001-arctrdhm003	5d 8h	N/A	96	1.5TB	N/A	N/A	N/A	For job steps requiring in excess of 32 CPUs or 768GB memory

GPU queue/partitions

Partitions	Nodes	Time limit	Priority	CPUs	Memory	GPUs	Limitations	Preemption
qTRDGPUH	arctrddgxa001-arctrddgxa003	5d 8h	high	40	512GB	gpu:V100:8	Max 8 GPUs per user	N/A
	arctrdgn001-arctrdgn002	5d 8h	high	40	192GB	gpu:V100:4	Max 8 GPUs per user	N/A
qTRDGPUM	arctrddgxa001-arctrddgxa003	5d 8h	medium	40	512GB	gpu:V100:8	Max 16 GPUs per user	suspend
	arctrdgn001-arctrdgn002	5d 8h	medium	40	192GB	gpu:V100:4	Max 16 GPUs per user	suspend
qTRDGPUL	arctrddgx001-arctrddgxa003	5d 8h	low	40	512GB	gpu:V100:8	N/A	suspend
	arctrdgn001-arctrdgn002	5d 8h	low	40	192GB	gpu:V100:4	N/A	suspend
qTRDGPU	arctrdagn001-arctrdagn019	5d 8h	N/A	64	512GB	gpu:RTX:1	N/A	N/A
	arctrdagn031-arctrdagn046	5d 8h	N/A	128	512GB	gpu:RTX:2	N/A	N/A

Special nodes

Nodes	CPUs	Memory	GPUs	Purpose
arclogin.rs.gsu.edu	0	1MB	N/A	Head node for submitting jobs on the cluster
arctrdcn017.rs.gsu.edu	32	768GB	N/A	Tasks that support research, such as docker and other services/daemons, VS Code remote development, scheduled task, data transfer etc.
arctrdcn018.rs.gsu.edu	32	768GB	N/A	Database server
arctrdgndev101.rs.gsu.edu	4	62GB	TITAN X:2	GPU development & testing
arctrdagn019.rs.gsu.edu	64	512GB	gpu:RTX:1	GPU development & testing
arctrdcn019-020.rs.gsu.edu	32	768GB	N/A	BrainForge CPU Queue
arctrdagn020.rs.gsu.edu	64	512GB	gpu:RTX:1	BrainForge GPU Queue

Using SLURM commands

# brief info and availability 
$ sinfo

# CPU and memory
$ sinfo -o "%24n %7P %.11T %.4c %.8m %14C %10e"

# CPU, memory and GPU
$ sinfo -O "nodehost:16,partition:.8,cpus:.8,memory:.8,cpusstate:.16,freemem:.8,gres:.16,gresused:.16,statelong:.8,time:.16"