Compute Platforms with GPUs

Production GPU Clusters

Zone Nodes Total Memory (GB) GPU Architecture Total GPUs GPUs per compute node Cores/ Node Total Cores Memory/ Node GPU peak performance (TFLOP/s double precision)sort descending GPU global memory (GB) Switch Peak TFLOPs (GPUs) Peak TFLOPS (CPUs+GPUs)
Max SCF 324 82,944 NVIDIA Kepler K20X 40 2 16 5,184 256 78,336 IB QDR 52.4 160.2
Corona CZ 170 43,520 1/2 AMD MI25 GPUs; 1/2 AMD MI60 GPUs 656 4 48 8,160 IB HDR 383.0
Pascal CZ 164 41,984 NVIDIA Tesla P100 (Pascal) 326 2 36 5,904 256 5.00 16 Omni-Path 1 727.8 1 926.1
Sierra SCF 4,474 1,382,400 NVIDIA V100 (Volta) 17,280 4 44 190,080 256 7.00 IB EDR 120 960.0 125 626.0
Lassen CZ 795 253,440 NVIDIA V100 (Volta) 2,664 4 44 34,848 256 7.00 IB EDR 22 192.2 23 047.2
RZAnsel RZ 62 17,280 NVIDIA V100 (Volta) 216 4 44 2,376 256 7.00 IB EDR 1 512.0 1 570.0
RZHasGPU RZ 20 2,560 NVIDIA Tesla K80 64 4 16 320 128 60.00 768 IB QDR 59.8 68.0
Surface CZ 162 41,500 NVIDIA Tesla K40 316 2 16 2,592 256 452.00 3,792 IB QDR 451.9 505.8
RZManta RZ 44 11,520 NVIDIA Tesla P100 144 4 20 880 256 677.00 2,304 IB EDR 676.8 693.6
Shark SCF 44 11,520 NVIDIA Tesla P100 144 4 20 880 256 677.00 2,304 IB EDR 676.8 693.6
Ray CZ 62 17,280 NVIDIA Tesla P100 216 4 20 1,240 256 1,015.00 3,456 IB EDR 1 015.2 1 040.4