用python异步编程模拟并发测量api调用lambda响应延迟

前置工作

1.建立lambda function
2.建立api gateway 并部署
api gateway的使用文档：https://docs.aws.amazon.com/zh_cn/apigateway/index.html
3.python写工具测量

代码实现

import requests,time
import asyncio
from aiohttp import ClientSession
import aiohttp
import matplotlib.pyplot as plt
url = "xxx"#填入自己函数调用api
tasks = []
run_time=[]
present_time=[]
start_time=[]
count=300
async def run():
    async with ClientSession() as session:
        async with session.get(url) as response:
            response = await response.read()
            present_time.append(time.time())
            print(response)

if __name__ == "__main__":
    loop = asyncio.get_event_loop()
    begin_time=time.time()
    for i in range(count):
        start_time.append(time.time())
        tasks.append(run())
    end_time = time.time()
    loop.run_until_complete(asyncio.wait(tasks))
    loop.close()
    end_time = time.time()
    print("运行时间：",str(end_time-begin_time))
    for i in range(count):
        run_time.append(present_time[i]-start_time[i])
    print(run_time)
    x=[i for i in range(count)]
    plt.plot(x,run_time)
    plt.show()

思考

为什么延迟随并发的数量增加而增加？
高并发如何实现？