Torch Hammer benchmark report

Sirius — Sirius A100 1-GPU stress (40GB) — compact-full_x3200c0s13b1n0_20261111_191308.csv
source: compact-full_x3200c0s13b1n0_20261111_191308.csv GPU: NVIDIA A100-SXM4-40GB nodes: 1 GPUs: 1 benchmarks: 8
nodes
1
GPUs
1
benchmarks
8
outliers
0

(a) Batched GEMM (float32) [TFLOP/s]

fleet mean: 18.9 TFLOP/s | mean power: 323W | max temp: 48 C
Fleet distribution of GPU mean performance (TFLOP/s)
0 0 0 0 1 GPU count 18.9–19.9 TFLOP/s: 1 GPUs 19.4 TFLOP/s mean
Fleet distribution of GPU mean power (W)
0 0 1 GPU count 323–324 W: 1 GPUs 324 W mean
Fleet distribution of GPU max temperature (°C)
0 0 1 GPU count 48.0–49.0 °C: 1 GPUs 48.5 °C mean
Fleet distribution of GPU mean SM utilization (%)
0 0 1 GPU count 100–101 %: 1 GPUs 100 % mean
Fleet distribution of GPU mean memory BW utilization (%)
0 0 1 GPU count 22.0–23.0 %: 1 GPUs 22.5 % mean
Fleet distribution of GPU mean GPU clock (MHz)
0 0 1 GPU count 1,410–1,411 MHz: 1 GPUs 1,410 MHz mean
node GPU 0 node mean vs fleet thrt
x3200c0s13b1n01919+0.0%

(b) Convolution (float32) [img/s]

fleet mean: 9,364,516.4 img/s | mean power: 390W | max temp: 52 C
Fleet distribution of GPU mean performance (img/s)
0 0 0 0 1 GPU count 9.4M–9.4M img/s: 1 GPUs 9.4M img/s mean
Fleet distribution of GPU mean power (W)
0 0 1 GPU count 390–391 W: 1 GPUs 390 W mean
Fleet distribution of GPU max temperature (°C)
0 0 1 GPU count 52.0–53.0 °C: 1 GPUs 52.5 °C mean
Fleet distribution of GPU mean SM utilization (%)
0 0 1 GPU count 99.0–100 %: 1 GPUs 99.5 % mean
Fleet distribution of GPU mean memory BW utilization (%)
0 0 1 GPU count 72.0–73.0 %: 1 GPUs 72.5 % mean
Fleet distribution of GPU mean GPU clock (MHz)
0 0 1 GPU count 1,395–1,396 MHz: 1 GPUs 1,396 MHz mean
node GPU 0 node mean vs fleet thrt
x3200c0s13b1n09,364,5169,364,516+0.0%

(c) Einsum Attention (float32) [TFLOP/s]

fleet mean: 17.1 TFLOP/s | mean power: 347W | max temp: 52 C
Fleet distribution of GPU mean performance (TFLOP/s)
0 0 0 0 1 GPU count 17.1–18.1 TFLOP/s: 1 GPUs 17.6 TFLOP/s mean
Fleet distribution of GPU mean power (W)
0 0 1 GPU count 347–348 W: 1 GPUs 348 W mean
Fleet distribution of GPU max temperature (°C)
0 0 1 GPU count 52.0–53.0 °C: 1 GPUs 52.5 °C mean
Fleet distribution of GPU mean SM utilization (%)
0 0 1 GPU count 100–101 %: 1 GPUs 100 % mean
Fleet distribution of GPU mean memory BW utilization (%)
0 0 1 GPU count 26.0–27.0 %: 1 GPUs 26.5 % mean
Fleet distribution of GPU mean GPU clock (MHz)
0 0 1 GPU count 1,410–1,411 MHz: 1 GPUs 1,410 MHz mean
node GPU 0 node mean vs fleet thrt
x3200c0s13b1n01717+0.0%

(d) Memory Traffic (float32) [GB/s]

fleet mean: 29.6 GB/s | mean power: 194W | max temp: 49 C
Fleet distribution of GPU mean performance (GB/s)
0 0 0 0 1 GPU count 29.6–30.6 GB/s: 1 GPUs 30.1 GB/s mean
Fleet distribution of GPU mean power (W)
0 0 1 GPU count 194–194 W: 1 GPUs 194 W mean
Fleet distribution of GPU max temperature (°C)
0 0 1 GPU count 49.0–50.0 °C: 1 GPUs 49.5 °C mean
Fleet distribution of GPU mean SM utilization (%)
0 0 1 GPU count 100–101 %: 1 GPUs 100 % mean
Fleet distribution of GPU mean memory BW utilization (%)
0 0 1 GPU count 51.0–52.0 %: 1 GPUs 51.5 % mean
Fleet distribution of GPU mean GPU clock (MHz)
0 0 1 GPU count 1,410–1,411 MHz: 1 GPUs 1,410 MHz mean
node GPU 0 node mean vs fleet thrt
x3200c0s13b1n03030+0.0%

(e) Heat Equation (float32) [MLUPS]

fleet mean: 15,501.0 MLUPS | mean power: 289W | max temp: 45 C
Fleet distribution of GPU mean performance (MLUPS)
0 0 0 0 1 GPU count 16k–16k MLUPS: 1 GPUs 16k MLUPS mean
Fleet distribution of GPU mean power (W)
0 0 1 GPU count 289–290 W: 1 GPUs 290 W mean
Fleet distribution of GPU max temperature (°C)
0 0 1 GPU count 45.0–46.0 °C: 1 GPUs 45.5 °C mean
Fleet distribution of GPU mean SM utilization (%)
0 0 1 GPU count 100–101 %: 1 GPUs 100 % mean
Fleet distribution of GPU mean memory BW utilization (%)
0 0 1 GPU count 100–101 %: 1 GPUs 100 % mean
Fleet distribution of GPU mean GPU clock (MHz)
0 0 1 GPU count 1,410–1,411 MHz: 1 GPUs 1,410 MHz mean
node GPU 0 node mean vs fleet thrt
x3200c0s13b1n015,50115,501+0.0%

(f) Schrödinger Equation (float32) [iter/s]

fleet mean: 58.8 iter/s | mean power: 201W | max temp: 45 C
Fleet distribution of GPU mean performance (iter/s)
0 0 0 0 1 GPU count 58.8–59.8 iter/s: 1 GPUs 59.3 iter/s mean
Fleet distribution of GPU mean power (W)
0 0 1 GPU count 201–202 W: 1 GPUs 202 W mean
Fleet distribution of GPU max temperature (°C)
0 0 1 GPU count 45.0–46.0 °C: 1 GPUs 45.5 °C mean
Fleet distribution of GPU mean SM utilization (%)
0 0 1 GPU count 99.0–100 %: 1 GPUs 99.5 % mean
Fleet distribution of GPU mean memory BW utilization (%)
0 0 1 GPU count 100–101 %: 1 GPUs 100 % mean
Fleet distribution of GPU mean GPU clock (MHz)
0 0 1 GPU count 1,410–1,411 MHz: 1 GPUs 1,410 MHz mean
node GPU 0 node mean vs fleet thrt
x3200c0s13b1n05959+0.0%

(g) Atomic Contention (float32) [Gops/s]

fleet mean: 9.2 Gops/s | mean power: 124W | max temp: 41 C
Fleet distribution of GPU mean performance (Gops/s)
0 0 0 0 1 GPU count 9.2–10.2 Gops/s: 1 GPUs 9.7 Gops/s mean
Fleet distribution of GPU mean power (W)
0 0 1 GPU count 124–125 W: 1 GPUs 125 W mean
Fleet distribution of GPU max temperature (°C)
0 0 1 GPU count 41.0–42.0 °C: 1 GPUs 41.5 °C mean
Fleet distribution of GPU mean SM utilization (%)
0 0 1 GPU count 99.0–100 %: 1 GPUs 99.5 % mean
Fleet distribution of GPU mean memory BW utilization (%)
0 0 1 GPU count 9.0–10.0 %: 1 GPUs 9.5 % mean
Fleet distribution of GPU mean GPU clock (MHz)
0 0 1 GPU count 1,410–1,411 MHz: 1 GPUs 1,410 MHz mean
node GPU 0 node mean vs fleet thrt
x3200c0s13b1n099+0.0%

(h) Sparse MM (float32) [GFLOP/s]

fleet mean: 242.7 GFLOP/s | mean power: 221W | max temp: 44 C
Fleet distribution of GPU mean performance (GFLOP/s)
0 0 0 0 1 GPU count 243–244 GFLOP/s: 1 GPUs 243 GFLOP/s mean
Fleet distribution of GPU mean power (W)
0 0 1 GPU count 221–222 W: 1 GPUs 222 W mean
Fleet distribution of GPU max temperature (°C)
0 0 1 GPU count 44.0–45.0 °C: 1 GPUs 44.5 °C mean
Fleet distribution of GPU mean SM utilization (%)
0 0 1 GPU count 100–101 %: 1 GPUs 100 % mean
Fleet distribution of GPU mean memory BW utilization (%)
0 0 1 GPU count 5.0–6.0 %: 1 GPUs 5.5 % mean
Fleet distribution of GPU mean GPU clock (MHz)
0 0 1 GPU count 1,410–1,411 MHz: 1 GPUs 1,410 MHz mean
node GPU 0 node mean vs fleet thrt
x3200c0s13b1n0243243+0.0%

Outliers

None detected