The pat cluster is a set of workstations belonging to David Wales which are configured with shared filesystems and a batch queueing system. Most of the workstations have GPUs. It is intended for running work within a single compute node. It should not be used for MPI between compute nodes as it does not have a high speed network for message passing.

It contains the following types of compute node

• Two nodes with Nvidia Tesla K20 GPUs, varying CPUs and RAM
• Thirty-four nodes with varying numbers (from 1 to 4) of Nvidia Titan Black GPUs, varying CPUs and RAM
• Eight nodes with no GPUs. These have twin 16-core Intel Xeon 6333P CPUs each and 128GB of RAM
• Four nodes with no GPUs. These have single 32-core Intel Xeon 6731P CPUs each and 128GB of RAM