mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-10-05 08:37:06 +08:00
65 lines
1.9 KiB
Python
65 lines
1.9 KiB
Python
#!/bin/env python3
|
||
# -*- coding: utf-8 -*-
|
||
# @author DDDivano
|
||
# encoding=utf-8 vi:ts=4:sw=4:expandtab:ft=python
|
||
|
||
|
||
import concurrent.futures
|
||
from collections import Counter
|
||
|
||
from core import TEMPLATE, URL, build_request_payload, send_request
|
||
|
||
|
||
def test_waiting_time():
|
||
"""
|
||
并发测试：
|
||
同时发起 1333 条请求。
|
||
校验所有返回码统计，且数量总和必须等于 1333。
|
||
额外校验：200 的数量必须小于 1333。
|
||
--max-num-seqs 128 \
|
||
--tensor-parallel-size 1 \
|
||
--max-concurrency 5000 \
|
||
--max-waiting-time 1 \
|
||
"""
|
||
|
||
data = {
|
||
"stream": False,
|
||
"messages": [
|
||
{"role": "user", "content": "1+1=？ 直接回答"},
|
||
],
|
||
"max_tokens": 10000,
|
||
"metadata": {
|
||
"min_tokens": 99,
|
||
},
|
||
"temperature": 0.8,
|
||
"top_p": 0,
|
||
}
|
||
|
||
def send_one_request(i):
|
||
payload = build_request_payload(TEMPLATE, data)
|
||
response = send_request(URL, payload)
|
||
print(f"请求 {i} 返回码: {response.status_code}")
|
||
return response.status_code
|
||
|
||
# 并发执行 1333 个请求
|
||
with concurrent.futures.ThreadPoolExecutor(max_workers=1333) as executor:
|
||
futures = [executor.submit(send_one_request, i) for i in range(1333)]
|
||
results = [f.result() for f in futures]
|
||
|
||
# 统计所有返回码
|
||
counter = Counter(results)
|
||
print("返回码统计结果:")
|
||
for code, cnt in sorted(counter.items()):
|
||
print(f" {code}: {cnt}")
|
||
|
||
# 校验返回总数
|
||
total = sum(counter.values())
|
||
assert total == 1333, f"返回数量不一致，总数={total}, 期望=1333"
|
||
|
||
# 校验 200 数量必须小于 1333
|
||
count_200 = counter.get(200, 0)
|
||
assert count_200 < 1333, f"200 数量错误，应小于1333，实际={count_200}"
|
||
assert count_200 >= 1024, f"200 数量错误，应大于等于1024，实际={count_200}"
|
||
|
||
print("并发请求校验通过")
|