47 lines
1.1 KiB
JSON
47 lines
1.1 KiB
JSON
{
|
|
"backend": "vllm-omni",
|
|
"task": "fl2va",
|
|
"gpus_per_instance": 4,
|
|
"dit_tensor_parallel_size": 2,
|
|
"ulysses_degree": 2,
|
|
"replicas": 2,
|
|
"expected_requests": 32,
|
|
"requests_recorded": 32,
|
|
"completed": 32,
|
|
"failed": 0,
|
|
"machine_wall_s": 3179.377814054489,
|
|
"machine_qps": 0.010064862332039778,
|
|
"latency_mean_s": 198.63877470575244,
|
|
"latency_p50_s": 142.95883168897126,
|
|
"latency_p95_s": 465.907953453809,
|
|
"by_short_edge": {
|
|
"480": {
|
|
"requests": 8,
|
|
"completed": 8,
|
|
"failed": 0,
|
|
"latency_mean_s": 43.13877707431675,
|
|
"latency_p95_s": 43.28898929227144
|
|
},
|
|
"720": {
|
|
"requests": 8,
|
|
"completed": 8,
|
|
"failed": 0,
|
|
"latency_mean_s": 133.32461226650048,
|
|
"latency_p95_s": 133.51795659495983
|
|
},
|
|
"768": {
|
|
"requests": 8,
|
|
"completed": 8,
|
|
"failed": 0,
|
|
"latency_mean_s": 152.63104947161628,
|
|
"latency_p95_s": 152.88410493800183
|
|
},
|
|
"1080": {
|
|
"requests": 8,
|
|
"completed": 8,
|
|
"failed": 0,
|
|
"latency_mean_s": 465.46066001057625,
|
|
"latency_p95_s": 466.80472294560167
|
|
}
|
|
}
|
|
} |