gil在CPU密集型多线程和单线程执行效率对比

直接用单线程执行for循环&＃xff0c;执行5次&＃xff0c;计算执行的时间

import logging
import datetimeFORMAT &＃61; &＃39;%(threadName)s %(thread)d %(message)s&＃39;
logging.basicConfig(format&＃61;FORMAT, level&＃61;logging.INFO)start &＃61; datetime.datetime.now()def calc():sum &＃61; 0for _ in range(100000000):sum &＃43;&＃61; 1calc()
calc()
calc()
calc()
calc()delta &＃61; (datetime.datetime.now() - start).total_seconds()
logging.info(delta)

用多线程的方式&＃xff0c;同样的for循环&＃xff0c;使用5个线程跑&＃xff0c;统计执行的时间

import threading
import logging
import datetimeFORMAT &＃61; &＃39;%(threadName)s %(thread)d %(message)s&＃39;
logging.basicConfig(format&＃61;FORMAT, level&＃61;logging.INFO)start &＃61; datetime.datetime.now()def calc():sum &＃61; 0for _ in range(100000000000):sum &＃43;&＃61; 1
lst &＃61; []
for i in range(5):t &＃61; threading.Thread(target&＃61;calc, )lst.append(t)t.start()for i in lst:i.join()delta &＃61; (datetime.datetime.now() - start).total_seconds()
logging.info(delta)

两个测试结果来看&＃xff0c;在CPython中&＃xff0c;对于CPU密集型而言&＃xff0c;多线程没特别大的优势&＃xff0c;和一个线程的执行时间相差不是特别大