NXG nodes

We have 4 NeXtScale nx360 M5 nodes on Apocrita with Nvidia Tesla K80 GPU cards.

NXG
Count	4
Processor	2 x 16 Core Intel Xeon E5-2683V3 (Broadwell)
Cores/Node	32
RAM	256GB
TMP Size	125GB
Interconnect	Mellanox ConnectX-4 EDR Infiniband
GPU	Tesla K80 with 2 x GK210 chip
GPU architecture	Kepler
CUDA cores	2496 per chip
GPU Memory	12GiB per chip
CUDA Compute	3.7

Accessing the nxg nodes

Access is permitted to QMUL researchers upon request. Note that access to GPU nodes is not permitted for Undergraduate and MSc students. Please raise a support ticket by emailing its-research-support@qmul.ac.uk with a brief overview of intended use, so that we can verify that the jobs will use the GPUs correctly. More detail is on the using GPUs section.

Half-node jobs

Please ensure that you specify system (as opposed to GPU) RAM carefully when requesting a "half-node job" (i.e. to make use of a single GPU). Remember that CPU memory is per core, so if you are using 1 GPU and 16 CPU cores, setting -l h_vmem=7.5G will request 120GB of system RAM.