apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2026-04-23 00:17:25 +08:00
Code Issues Actions 19 Packages Projects Releases Wiki Activity

Revert "[Optimization] Optimize ttft for prefill pd (#6680)" (#7386)

Browse Source 
* Revert "[Optimization] Optimize ttft for prefill pd (#6680)"

This reverts commit 6727df8286.

* fix revert pr
This commit is contained in:
freeliuzc
2026-04-14 20:01:39 +08:00
committed by GitHub
parent 8a8beca548
commit f6c066fb9d
11 changed files with 135 additions and 134 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/ci_use/metrics/test_metrics.py
+16 -17
View File
@@ -214,29 +214,28 @@ def test_metrics_with_clear_and_reset():
"""
Test the metrics monitoring endpoint.
"""
pass # not stable, uncomment after bug fix
# metrics_url = f"http://0.0.0.0:{FD_METRICS_PORT}/metrics"
metrics_url = f"http://0.0.0.0:{FD_METRICS_PORT}/metrics"
# async_concurrency(n=10)
async_concurrency(n=10)
# time.sleep(0.3)
time.sleep(0.3)
# ===== clear_load_weight =====
# clear_url = f"http://0.0.0.0:{FD_API_PORT}/clear_load_weight"
# print("Calling clear_load_weight...")
# r = requests.get(clear_url, timeout=30)
# assert r.status_code == 200, f"clear_load_weight failed: {r.status_code}"
clear_url = f"http://0.0.0.0:{FD_API_PORT}/clear_load_weight"
print("Calling clear_load_weight...")
r = requests.get(clear_url, timeout=30)
assert r.status_code == 200, f"clear_load_weight failed: {r.status_code}"
# metrics = get_metrics_dict(metrics_url)
# running = metrics["fastdeploy:num_requests_running"]
# waiting = metrics["fastdeploy:num_requests_waiting"]
metrics = get_metrics_dict(metrics_url)
running = metrics["fastdeploy:num_requests_running"]
waiting = metrics["fastdeploy:num_requests_waiting"]
# print(
# "ASSERT after the clear_load_weight operation, the value is 0 (Request interruption stopped inference, and related requests were cleared):",
# running,
# "waiting:",
# waiting,
# )
print(
"ASSERT after the clear_load_weight operation, the value is 0 (Request interruption stopped inference, and related requests were cleared):",
running,
"waiting:",
waiting,
)
# assert running == 0 and waiting == 0, "Expected both running and waiting to be 0 after clear_load_weight"