We have 12 new GPU node added to the cluster. Each node has two NVIDIA Tesla P100 GPU cards. These nodes (Nucleus 162-173) are in a new GPUv1 partition, since they have improved specification than the previous 8 Tesla K20/K40 GPUs.
To run a job on the new GPU nodes, please specify ‘GPUv1’ in the SLURM script. You can also access the nodes through portal website – cloud services - web visualization, and choose ‘webGPUv1’.
** IMPORTANT **: To run applications using the P100 GPUs, you DO NOT NEED to add vglrun before your applications on these nodes. For Nucleus042-049, you still need vglrun.