GPU nodes for specific Computer Science academics

Four academics in the Department of Computer Science (DCS) share two NVIDIA V100 GPU nodes in Bessemer:

Academic	Node	Slurm Account name	Slurm Partition name
Carolina Scarton	`bessemer-node041`	`dcs-acad1`	(see notes below)
Chenghua Lin	`bessemer-node041`	`dcs-acad2`	(see notes below)
Matt Ellis	`bessemer-node042`	`dcs-acad3`	(see notes below)
Po Yang	`bessemer-node042`	`dcs-acad4`	(see notes below)

Hardware specifications

`bessemer-node041` and `bessemer-node042`

Processors	2x Intel Xeon Gold 6138 (2.00GHz; 20 cores per CPU)
RAM	192GB (DDR4 @ 2666 MHz)
NUMA nodes	2x
GPUS	4x NVIDIA Tesla V100 SXM2 (16GB RAM each; NVLINK interconnects between GPUs)
Networking	25 Gbps Ethernet
Local storage	140 GB of temporary storage under `/scratch` (2x SSD RAID1)

Requesting access

Users other than the listed academics should contact one of those academics should they want access to these nodes.

That academic can then grant users access to the relevant SLURM Account (e.g. dcs-acad1) via this web interface.

Using the nodes

There are several ways to access these nodes. The type of access granted for a job depends on which SLURM Account and Partition are requested at job submission time. Only certain users have access to a given Account.

bessemer-node041 and bessemer-node042: non-preemptable access to half a node

Each of four academics (plus their collaborators) have ring-fenced, on-demand access to the resources of half a node.

To submit a job via this route, you need to specify a Partition and an Account when submitting a batch job or starting an interactive session:

Partition: dcs-acad
Account: dcs-acadX where X is 1, 2, 3 or 4 and varies between the academics).
QoS: do not specify one i.e. do not use the --qos parameter.

Resource limits per job:

Default run-time: 8 hours
Maximum run-time: 7 days
CPU cores: 20
GPUs: 2
Memory: 96 GB

bessemer-node041 and bessemer-node042: preemptable access to both nodes

If any of the four academics (or their collaborators) want to run a larger job that requires up to all the resources available in one of these two nodes then they can specify a different Partition when submitting a batch job or starting an interactive session:

Partition: dcs-acad-pre
Account: dcs-acadX where X is 1, 2, 3 or 4 and varies between the academics).
QoS: do not specify one i.e. do not use the --qos parameter.

However, to facilitate fair sharing of these GPU nodes jobs submitted via this route are preemptable: they will be stopped mid-execution if a job is submitted to the dcs-acad partition (see above) that requires those resources.

When a job submitted by this route is preempted by another job the preempted job is terminated and re-queued.