High Performance Computing

• High Performance Computing
• Progetti TLC
• Spatial data, Projects and Services
• Valorization and Transfer of Knowledge

CRS4 offers support for research to the scientific community also through the High Performance Computing, thanks to a constantly updated computing center. The HPC group supports researchers and users, also in the industrial sector, providing adequate services that will suit all their needs. CRS4 offers more than 400 compute nodes divided in different clusters: in total more than 5500 computational cores availables.

• Dell cluster, 16 compute nodes with Infiniband EDR 100 Gbps
• Huawei cluster, 32 nodi with Infiniband FDR 56 Gbps
• Join Innovation Center (JIC) cluster, a total of 104 compute nodes with Infiniband FDR 56 Gbps and 10 Gbps Ethernet; this cluster groups hpc compute nodes, Big Data nodes and GPU compute nodes (with Nvidia Kepler K40)
• Dell GPU cluster, 10 nodi with Infiniband FDR 56 Gbps and GPU AMD firepro
• Supermicro hybrid cluster, 10 nodi with Infiniband QDR 40 Gbps and dual Nvidia Kepler K10 or Intel Phi accelerator
• Supermicro GPU cluster, 6 nodi with Infiniband FDR 56 Gbps with GPU Nvidia Kepler K40
• HP Cluster, 256 nodi with Infiniband DDR 20 Gbps
• Access and interactive nodes, graphical nodes and service nodes

Our newer most significant HPC acquisitions is a 16 nodes Dell cluster, based on Sky Lake architecture Xeon Gold 6142, with Infiniband EDR 100 Gbps and 10 Gbps Ethernet.

---