Welcome to the Crimson GPU Cluster

The Crimson GPU cluster is a group of servers where users submit their intensive GPU workloads to be executed.

Benefits of having a GPU

When code is ran on a computer, the computer's CPU does all the calculations and heavy-lifting to get the results required. Having a GPU available can provide support to the CPU by performing concurrent calculations. This allows the GPU to complete simple and repetitive tasks much faster as it breaks down these tasks into smaller componenets and finish them in parallel.

To understand better, consider the following analogy:

The CPU is like a head chef in a large restaurant who has to make sure hundreds of burgers get flipped. Even if the head chef can do it personally, it’s not the best use of time. All kitchen operations may halt or slow down while the head chef is completing this simple but time-consuming task. To avoid this, the head chef can use junior assistants who flip several burgers in parallel. The GPU is more like a junior assistant with ten hands who can flip 100 burgers in 10 seconds.

Utilising the GPU cluster

Users will submit their jobs (code to be executed) via the login node, and depending on the resource availability. The job may be either executed immediately or be placed in a queue and executed at a later time when resources are available.

The GPU Cluster is available only on the SMU Network. If you are connecting remotely, a SMU VPN is required. Contact your instructor or professor for a SMU VPN account.

Resources available

Please refer to the Google Spreadsheet (here) for the list of resources available on the cluster. Log in with your SMU credentials (abc@smu.edu.sg) in order to view the spreadsheet.