kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-02-04 18:21:52 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

[None][test] store per user output and per gpu output metric in csv file (#10658)

Browse Source 
Signed-off-by: Ruodi Lu <ruodil@users.noreply.github.com>
Co-authored-by: Ruodi Lu <ruodil@users.noreply.github.com>
This commit is contained in:
ruodil 2026-01-15 13:51:08 +08:00 committed by GitHub
parent 7b3b6f1161
commit 22240e43eb
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 9 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
tests/integration/defs/perf/test_perf.py
View File

@ -342,6 +342,11 @@ BENCH_PERF_METRIC_LOG_QUERIES = {
PerfMetricType.KV_CACHE_SIZE:
re.compile(r".*(?:Allocated ([\d\.]+) GiB for max tokens in paged KV cache|"
r"Final KV cache size after resize: ([\d\.]+) GiB).*"),
PerfMetricType.PER_USER_OUTPUT_THROUGHPUT:
re.compile(
r"Per User Output Throughput \[w\/ ctx\] \(tps\/user\):\s+([\d\.]+)"),
PerfMetricType.PER_GPU_OUTPUT_THROUGHPUT:
re.compile(r"Per GPU Output Throughput \(tps\/gpu\):\s+([\d\.]+)"),
}
AGGR_SERVER_PERF_METRIC_LOG_QUERIES = {
@ -457,6 +462,8 @@ PERF_METRIC_STRING = {
PerfMetricType.ENGINE_SIZE: "engine_size",
PerfMetricType.CONTEXT_GPU_MEMORY: "context_gpu_memory",
PerfMetricType.KV_CACHE_SIZE: "kv_cache_size",
PerfMetricType.PER_USER_OUTPUT_THROUGHPUT: "per_user_output_throughput",
PerfMetricType.PER_GPU_OUTPUT_THROUGHPUT: "per_gpu_output_throughput",
}
BUILDER_METRICS = [

2
tests/integration/defs/perf/utils.py
View File

@ -114,6 +114,8 @@ class PerfMetricType(str, Enum):
KV_CACHE_SIZE = "KV_CACHE_SIZE"
DISAGG_SERVER_E2EL = "DISAGG_SERVER_E2EL"
DISAGG_SERVER_TTFT = "DISAGG_SERVER_TTFT"
PER_USER_OUTPUT_THROUGHPUT = "PER_USER_OUTPUT_THROUGHPUT"
PER_GPU_OUTPUT_THROUGHPUT = "PER_GPU_OUTPUT_THROUGHPUT"
@contextlib.contextmanager