Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

feat: add e2e latency #704

Merged
merged 2 commits into from
Jul 23, 2024
Merged

feat: add e2e latency #704

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c5c17ab
feat: add e2e latency
zhyncs Jul 23, 2024
c0a5768
update
zhyncs Jul 23, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 21 additions & 0 deletions python/sglang/bench_serving.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -264,6 +264,8 @@ class BenchmarkMetrics:
median_itl_ms: float
std_itl_ms: float
p99_itl_ms: float
mean_e2e_latency_ms: float
median_e2e_latency_ms: float


default_sharegpt_path = "ShareGPT_V3_unfiltered_cleaned_split.json"
Expand Down Expand Up @@ -467,6 +469,7 @@ def calculate_metrics(
itls: List[float] = []
tpots: List[float] = []
ttfts: List[float] = []
e2e_latencies: List[float] = []
for i in range(len(outputs)):
if outputs[i].success:
output_len = outputs[i].output_len
Expand All @@ -480,6 +483,9 @@ def calculate_metrics(
tpots.append((outputs[i].latency - outputs[i].ttft) / (output_len - 1))
itls += outputs[i].itl
ttfts.append(outputs[i].ttft)

e2e_latencies.append(outputs[i].latency)

completed += 1
else:
output_lens.append(0)
Expand Down Expand Up @@ -513,6 +519,8 @@ def calculate_metrics(
median_itl_ms=np.median(itls or 0) * 1000,
std_itl_ms=np.std(itls or 0) * 1000,
p99_itl_ms=np.percentile(itls or 0, 99) * 1000,
mean_e2e_latency_ms=np.mean(e2e_latencies) * 1000,
median_e2e_latency_ms=np.median(e2e_latencies) * 1000,
)

return metrics, output_lens
Expand Down Expand Up @@ -611,6 +619,15 @@ async def benchmark(
"Output token throughput (tok/s):", metrics.output_throughput
)
)
print("{s:{c}^{n}}".format(s="End-to-End Latency", n=50, c="-"))
print(
"{:<40} {:<10.2f}".format("Mean E2E Latency (ms):", metrics.mean_e2e_latency_ms)
)
print(
"{:<40} {:<10.2f}".format(
"Median E2E Latency (ms):", metrics.median_e2e_latency_ms
)
)
print("{s:{c}^{n}}".format(s="Time to First Token", n=50, c="-"))
print("{:<40} {:<10.2f}".format("Mean TTFT (ms):", metrics.mean_ttft_ms))
print("{:<40} {:<10.2f}".format("Median TTFT (ms):", metrics.median_ttft_ms))
Expand Down Expand Up @@ -639,6 +656,8 @@ async def benchmark(
"total_input": metrics.total_input,
"total_output": metrics.total_output,
"total_output_retokenized": metrics.total_output_retokenized,
"mean_e2e_latency": metrics.mean_e2e_latency_ms,
"median_e2e_latency": metrics.median_e2e_latency_ms,
"median_ttft": metrics.median_ttft_ms,
"median_itl": metrics.median_itl_ms,
"output_token_throughput": metrics.output_throughput,
Expand Down Expand Up @@ -693,6 +712,8 @@ async def benchmark(
"itls": [output.itl for output in outputs],
"generated_texts": [output.generated_text for output in outputs],
"errors": [output.error for output in outputs],
"mean_e2e_latency_ms": metrics.mean_e2e_latency_ms,
"median_e2e_latency_ms": metrics.median_e2e_latency_ms,
}
return result

Expand Down
Loading