test_gpu_scripts/configs/default.yaml

79 lines
2.0 KiB
YAML
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

# GPU type: auto-detect or override to a100/a800/h100/h800/h200/h20/b200/b300
gpu_type: auto
benchmark:
memory:
size_mb: 4096
iterations: 10
nvbandwidth_buffer_mb: 512
nvbandwidth_samples: 3
compute:
dtypes:
- fp32
- tf32
- fp16
- bf16
- fp8
# MAMF-style shape sweep: measure each dtype at every shape below and keep the max
# TFLOPS (the realistic achievable peak). A single fixed shape under-reports by
# ~7-12% and can't meet the MAMF-calibrated thresholds in gpu_specs.py.
# Each entry is either N (square N×N×N) or [M, N, K]. K-heavy non-square shapes
# (e.g. 2048×2048×13312) hit the true Hopper MAMF — bf16 ~790 vs ~755 square.
# Empty list => single matrix_size shape (legacy behaviour).
sweep_sizes:
- 3584
- 4608
- 5376
- 8192
- 11520
- [2048, 2048, 13312]
- [2048, 2048, 16384]
matrix_size: 8192 # fallback shape when sweep_sizes is empty
warmup: 20
iterations: 80
# NOTE: torch.compile was dropped — on H100 eager cuBLAS beats Triton for plain
# GEMM, and compiling would re-autotune per shape and make the sweep very slow.
health:
temp_warning: 75
temp_critical: 85
power_limit: null # null = auto-detect from GPU TDP per gpu_specs.py
nccl:
min_bandwidth_gbps: null # null = auto-detect (40% of GPU NVLink BW)
test_allreduce: true
test_alltoall: true
test_broadcast: true
test_reduce_scatter: false
test_allgather: false
test_sendrecv: false
stress:
duration_sec: 600 # 10 min — reaches thermal steady state, validates throttle/jitter beyond warmup
use_doubles: false
use_tensor_cores: true
memory_pct: 90
gpus: all
rdma:
min_bandwidth_gbps: 50
max_latency_us: 10
ib_iterations: 1000
msg_size: 65536
ib_device: null
ib_port: 1
training:
model: gpt2
batch_size: 8
seq_length: 2048
num_steps: 50
dtype: bf16
report:
output_dir: ./reports
format: json
tools:
install_dir: /opt/gpu-test-tools