GPU Cluster

The High-Performance Computing (HPC) cluster with General Purpose Graphical Processing Units (GPGPU) was established in 2019, equipped with a total of 20,480 CUDA cores, 2,560 Tensor cores and 128GB graphical memory, providing a theoretical peak deep learning capacity at 112TFLOPS. 

How to Participate in the Cluster Project?

The cluster management team is looking for cooperative projects to improve clinical diagnostic and cluster management skills. If you are interested in the cluster, please contact the department office via radiology@cuhk.edu.hk with attention to Matthew Wong.

Supporting Libraries

Researchers are encouraged to install any software and library in the cluster necessary for their work. The libraries currently supported by our system include but are not limited to the following: