HSE University HPC Cluster "cHARISMa"

At the beginning of 2019, a high-performance cluster was launched at HSE University. The cluster placed 10^th on the Top-50 ranking of CIS high-performance computer systems.

The high-performance cluster consists of 46 computing nodes, including 6 newest nodes for training neural networks with eight NVIDIA A100 80GB SXM GPUs each, 29 specialized nodes with a large amount of RAM 768-1536 GB and four NVIDIA Tesla V100 32 GB graphics accelerators each, as well as 11 computing nodes with powerful central processors.

The resources of the computing cluster are intended to support basic research and teaching at university, as well as to carry out research projects requiring the use of high-performance systems. Cluster resources are allocated upon request for specific projects and for a limited period of time.

Currently, the HSE HPC cluster is ranked 10th in the Top50 ranking of the most powerful computer systems in CIS.

Types of computing nodes

The HSE supercomputer uses seven types of high-performance computing nodes. Type B nodes are used for tasks with very large datasets. Type C nodes provide tighter CPU-GPU connectivity than Types A and B. Type D nodes are optimal for CPU-centric workloads. Type E nodes are preferable for deep neural network training and large-scale data processing. Type F nodes are universal GPU nodes for computational experiments, as well as training and inference of small and medium-sized models. Type H nodes are intended for large-scale distributed training, including large language and multimodal models, and other resource-intensive workloads. See more about choosing the type of computing nodes to run a job.