Chair of Data Science HPC Cluster
The Chair of Data Science runs a High Performance Computing Cluster on the premises of the University of Würzburg. The cluster allows distributed computations on large datasets for research purposes and big data applications. The considerable number of GPUs enable large scale machine learning experiments which are necessary for modern AI research.
Workloads are scheduled in containers through Kubernetes with Ceph providing a distributed filesystem. We also run Hadoop, HBase, and Accumulo within the cluster.
General Information
| Nodes | 23 |
| Physical CPU cores | 640 |
| Logical CPU cores | 1280 |
| GPUs | 45 |
| System Memory | 5824 GB |
| GPU Memory | 498 GB |
| Total Memory | 6322 GB |
| Theoretical Perfomance (FP32) | 580 TFLOPS |
| Distributed Storage Capacity | 575 TB |