Note: The clusters receive regular maintenance and upgrades, so at times these CPU and GPU numbers may lag behind the latest changes. To see the most up-to-date information, SSH into your cluster of interest and run the "snodes" or "shownodes" commands. To see more detailed information, connect to a compute node and run "lscpu", "lsmem", and for GPUs "nvidia-smi -a".ClusterProcessorSpeedNodesCoresper NodeMemoryper NodeTotal CoresInterconnectTheoreticalFP64 PerformanceNobelDell Linux Server2.6 GHz Intel Ice Lake2641 TB128N/A10.6 TFLOPSAdroitDell Linux Cluster2.6 GHz Intel Skylake932384 GB288FDR Infiniband24 TFLOPSAdroit (adroit-h11g1)GPU node2.6 GHz Intel Sapphire Rapids1.41 GHz A100 PCIe1484 GPU/node770 GB80 GB/GPU48N/A3.3 TFLOPS28 TFLOPSAdroit (adroit-h11g2)GPU node2.8 GHz Intel Ice Lake1.41 GHz A1001488 GPU/node1 TB20 GB/GPU48N/A4.3 TFLOPS39 TFLOPSAdroit (adroit-h11g3)GPU node2.6 GHz Intel Skylake1230 MHz V1001564 GPU/node770 GB32 GB/GPU56N/A3.3 TFLOPS28 TFLOPSDellaDell Linux Cluster2.4 GHz AMD EPYC 96542.8 GHz Intel Cascade Lake5564192321500 GB190 GB12608HDR Infiniband FDR Infiniband995 TFLOPSDellaGPU nodes2.1 GHz Intel SR 8468H100 SXM42968 GPU/node1000 GB80 GB/GPU4032HDR100 Infiniband11.7 PFLOPSDellaGPU nodes2.8 GHz Intel Ice Lake1.41 GHz A10070484 GPU/node1 TB80 GB/GPU3360HDR100 Infiniband301 TFLOPS2716 TFLOPSDellaGPU nodes2.6 GHz AMD EPYC Rome1.41 GHz A100201282 GPU/node768 GB40 GB/GPU2560HDR100 Infiniband106 TFLOPS388 TFLOPSTigerCPU Nodes2.0 GHz Intel Sapphire Rapids3201121 TB35840NDR Infiniband2.3 PFLOPSTigerGPU Nodes2.0 GHz Intel Sapphire Rapids1.98 GHz H100121124 GPU/node1 TB80 GB/GPU1344NDR Infiniband1.6 PFLOPSStellarIntel nodes2.9 GHz Intel Cascade Lake296961 TB28416HDR100 Infiniband2600 TFLOPSStellarAMD nodes2.6 GHz AMD EPYC Rome187128512 GB23936HDR100 Infiniband996 TFLOPSStellarGPU nodes2.6 GHz AMD EPYC Rome1.41 GHz A10061282 GPU/node500 GB40 GB/GPU768HDR100 Infiniband32 TFLOPS116 TFLOPSThe A100 has a FP64 performance of 9.7 TFLOPS while the H100 SXM provides 34 TFLOPS.
