Accessing GPU nodes

Oscar has 44 GPU nodes that are regular compute nodes with two NVIDIA Tesla M2050 GPUs (Fermi architecture) added. Each M2050 GPU has 448 CUDA cores and 3GB GDDR5 memory. To gain access to these nodes, please submit a support ticket and ask to be added to the 'gpu' group.

Interactive Use

The gpu001 and gpu002 nodes are available for interactive logins. From an Oscar login node, use:

ssh gpu001   OR
ssh gpu002

to connect and run programs on the GPUs interactively. These nodes are intended for debugging and testing and their resources are shared among all users who are logged in.

GPU Queue

For production runs with exclusive access to GPU nodes, please submit a batch job to the gpu queue using:

qsub -q gpu <batch script>

You can see the status of the gpu queue with:

allq gpu