Python多线程多进程实例对比解析

多线程适合于多io操作

多进程适合于耗cpu(计算)的操作

# 多进程编程
# 耗cpu的操作,用多进程编程, 对于io操作来说,使用多线程编程
import time
from concurrent.futures import ThreadPoolExecutor, as_completed
from concurrent.futures import ProcessPoolExecutor

def fib(n):
 if n <= 2:
   return 1
 return fib(n - 2) + fib(n - 1)

if __name__ == '__main__':

# 1. 对于耗cpu操作,多进程优于多线程

# with ThreadPoolExecutor(3) as executor:
 #   all_task = [executor.submit(fib, num) for num in range(25, 35)]
 #   start_time = time.time()
 #   for future in as_completed(all_task):
 #     data = future.result()
 #     print(data)
 #   print("last time :{}".format(time.time() - start_time)) # 3.905290126800537

# 多进程 ,在window环境 下必须放在main方法中执行,否则抛异常
 with ProcessPoolExecutor(3) as executor:
   all_task = [executor.submit(fib, num) for num in range(25, 35)]
   start_time = time.time()
   for future in as_completed(all_task):
     data = future.result()
     print(data)
   print("last time :{}".format(time.time() - start_time)) # 2.6130592823028564

可以看到在耗cpu的应用中,多进程明显优于多线程 2.6130592823028564 < 3.905290126800537

下面模拟一个io操作

# 多进程编程
# 耗cpu的操作,用多进程编程, 对于io操作来说,使用多线程编程
import time
from concurrent.futures import ThreadPoolExecutor, as_completed
from concurrent.futures import ProcessPoolExecutor

def io_operation(n):
 time.sleep(2)
 return n

if __name__ == '__main__':

# 1. 对于耗cpu操作,多进程优于多线程

# with ThreadPoolExecutor(3) as executor:
 #   all_task = [executor.submit(io_operation, num) for num in range(25, 35)]
 #   start_time = time.time()
 #   for future in as_completed(all_task):
 #     data = future.result()
 #     print(data)
 #   print("last time :{}".format(time.time() - start_time)) # 8.00358772277832

# 多进程 ,在window环境 下必须放在main方法中执行,否则抛异常
 with ProcessPoolExecutor(3) as executor:
   all_task = [executor.submit(io_operation, num) for num in range(25, 35)]
   start_time = time.time()
   for future in as_completed(all_task):
     data = future.result()
     print(data)
   print("last time :{}".format(time.time() - start_time)) # 8.12435245513916

可以看到 8.00358772277832 < 8.12435245513916, 即是多线程比多进程更牛逼!

来源：https://www.cnblogs.com/z-qinfeng/p/12064012.html