[CI/Build] Resolve hardware-nested perf baselines per concurrency sweep (#5845)

Signed-off-by: wangyu <410167048@qq.com>
Signed-off-by: [Your Name] <your.email@example.com>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
This commit is contained in:
wangyuandamy-why-3459 authored and GitHub committed 2026-08-06 21:20:04 +08:00
1 parent 48298030d4
commit 4b625d1175
15 files changed
+414 -106

No files matched your search

+4 -2
View File
@@ -237,7 +237,7 @@ def test_performance_benchmark(omni_server, benchmark_params):
break
# QPS / request-rate sweep
for qps, num_prompt in zip(qps_list, num_prompt_list):
for sweep_index, (qps, num_prompt) in enumerate(zip(qps_list, num_prompt_list)):
args = args + ["--request-rate", str(qps), "--num-prompts", str(num_prompt)]
result = run_benchmark(
args=args,
@@ -246,6 +246,7 @@ def test_performance_benchmark(omni_server, benchmark_params):
dataset_name=dataset_name,
num_prompt=num_prompt,
baseline_config=params.get("baseline"),
sweep_index=sweep_index,
random_input_len=params.get("random_input_len"),
random_output_len=params.get("random_output_len"),
resource_label=resource_label,
@@ -253,7 +254,7 @@ def test_performance_benchmark(omni_server, benchmark_params):
assert_result(result, num_prompt)
# concurrency test
for concurrency, num_prompt in zip(max_concurrency_list, num_prompt_list):
for sweep_index, (concurrency, num_prompt) in enumerate(zip(max_concurrency_list, num_prompt_list)):
args = args + ["--max-concurrency", str(concurrency), "--num-prompts", str(num_prompt), "--request-rate", "inf"]
result = run_benchmark(
args=args,
@@ -262,6 +263,7 @@ def test_performance_benchmark(omni_server, benchmark_params):
dataset_name=dataset_name,
num_prompt=num_prompt,
baseline_config=params.get("baseline"),
sweep_index=sweep_index,
random_input_len=params.get("random_input_len"),
random_output_len=params.get("random_output_len"),
resource_label=resource_label,