The High-Performance Computing (HPC) cluster is a heterogeneous system because this system contains with and without GPU accelerators. Over time, there will be different architectures and generations of both the CPUs and GPUs.
The HPC cluster consists of the following infrastructure;
7x Compute Nodes (CPU Queue)
- CPU: 2x Xeon Gold 6230R Cascade Lake (26 core, 2.1GHz)
- RAM: 128GB Each
2x GPU Accelerated Nodes (GPU Queue)
- GPU: Nvidia V100 (32 GB, 5120 CUDA cores)
- CPU: 2x Xeon Gold 6230R Cascade Lake (26 core, 2.1GHz)
- RAM: 128GB Each
Master Node (Login Node)
- CPU: 1x CPU with 20 cores
- RAM: 47GB