dsr1-fp8-1k1k-max-tpt

concurrency 1,024 · 10 days ago

cron

passed

re-run

Commit

SHAa3fd5c2

feat: add NVTX markers for the scheduler main loop (#27901)

JoyFuture·10 days ago

PR #27901 · feat: add NVTX markers for the scheduler main loop

Run

GitHub Actions27486985911-1

Slurm job5305

GPUs48·prefill 16 / decode 32

ISL / OSL1024 / 1024

Logscron/27486985911-1/1k1k/dsr1-fp8-1k1k-max-tpt/2026-06-14/5305/sa-bench_isl_1024_osl_1024/

Metrics

28 captured

best_of

1.00

burstiness

1.00

completed

10,240

duration

259

max_concurrency

1,024

mean_e2el_ms

24,851ms

mean_itl_ms

1,238ms

mean_tpot_ms

25.34ms

mean_ttft_ms

1,516ms

median_e2el_ms

24,671ms

median_itl_ms

1,274ms

median_tpot_ms

25.48ms

median_ttft_ms

945ms

num_prompts

10,240

output_throughput

36,407tok/s

p99_e2el_ms

31,519ms

p99_itl_ms

1,567ms

p99_tpot_ms

26.56ms

p99_ttft_ms

7,356ms

peak_output_tokens_per_s

43,089s

request_throughput

39.49tok/s

std_e2el_ms

2,104ms

std_itl_ms

193ms

std_tpot_ms

0.79ms

std_ttft_ms

1,690ms

total_input_tokens

9,440,068

total_output_tokens

9,440,624

total_token_throughput

72,812tok/s