手记

Python中单线程、多线程和多进程的效率对比实验

           

原文出处: 饒木陽   

Python是运行在解释器中的语言,查找资料知道,python中有一个全局锁(GIL),在使用多线程(Thread)的情况下,不能发挥多核的优势。而使用多进程(Multiprocess),则可以发挥多核的优势真正地提高效率。

对比实验

资料显示,如果多线程的进程是CPU密集型的,那多线程并不能有多少效率上的提升,相反还可能会因为线程的频繁切换,导致效率下降,推荐使用多进程;如果是IO密集型,多线程进程可以利用IO阻塞等待时的空闲时间执行其他线程,提升效率。所以我们根据实验对比不同场景的效率

操作系统CPU内存硬盘
Windows 10双核8GB机械硬盘
(1)引入所需要的模块


1

2

3

4

import requests

import time

from threading import Thread

from multiprocessing import Process


(2)定义CPU密集的计算函数


Python

1

2

3

4

5

6

7

def count(x, y):

    # 使程序完成150万计算

    c = 0

    while c < 500000:

        c += 1

        x += x

        y += y


(3)定义IO密集的文件读写函数


1

2

3

4

5

6

7

8

9

10

def write():

    f = open("test.txt", "w")

    for x in range(5000000):

        f.write("testwrite\n")

    f.close()

 

def read():

    f = open("test.txt", "r")

    lines = f.readlines()

    f.close()


(4) 定义网络请求函数


1

2

3

4

5

6

7

8

9

10

_head = {

            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/48.0.2564.116 Safari/537.36'}

url = "http://www.tieba.com"

def http_request():

    try:

        webPage = requests.get(url, headers=_head)

        html = webPage.text

        return {"context": html}

    except Exception as e:

        return {"error": e}


(5)测试线性执行IO密集操作、CPU密集操作所需时间、网络请求密集型操作所需时间


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

# CPU密集操作

t = time.time()

for x in range(10):

    count(1, 1)

print("Line cpu", time.time() - t)

 

# IO密集操作

t = time.time()

for x in range(10):

    write()

    read()

print("Line IO", time.time() - t)

 

# 网络请求密集型操作

t = time.time()

for x in range(10):

    http_request()

print("Line Http Request", time.time() - t)


输出

  • CPU密集:95.6059999466、91.57099986076355 92.52800011634827、 99.96799993515015

  • IO密集:24.25、21.76699995994568、21.769999980926514、22.060999870300293

  • 网络请求密集型: 4.519999980926514、8.563999891281128、4.371000051498413、4.522000074386597、14.671000003814697

(6)测试多线程并发执行CPU密集操作所需时间


Python

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

counts = []

t = time.time()

for x in range(10):

    thread = Thread(target=count, args=(1,1))

    counts.append(thread)

    thread.start()

 

e = counts.__len__()

while True:

    for th in counts:

        if not th.is_alive():

            e -= 1

    if e <= 0:

        break

print(time.time() - t)


Output: 99.9240000248 、101.26400017738342、102.32200002670288

(7)测试多线程并发执行IO密集操作所需时间


Python

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

def io():

    write()

    read()

 

t = time.time()

ios = []

t = time.time()

for x in range(10):

    thread = Thread(target=count, args=(1,1))

    ios.append(thread)

    thread.start()

 

e = ios.__len__()

while True:

    for th in ios:

        if not th.is_alive():

            e -= 1

    if e <= 0:

        break

print(time.time() - t)


Output: 25.69700002670288、24.02400016784668

(8)测试多线程并发执行网络密集操作所需时间


Python

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

t = time.time()

ios = []

t = time.time()

for x in range(10):

    thread = Thread(target=http_request)

    ios.append(thread)

    thread.start()

 

e = ios.__len__()

while True:

    for th in ios:

        if not th.is_alive():

            e -= 1

    if e <= 0:

        break

print("Thread Http Request", time.time() - t)


Output: 0.7419998645782471、0.3839998245239258、0.3900001049041748

(9)测试多进程并发执行CPU密集操作所需时间


Python

1

2

3

4

5

6

7

8

9

10

11

12

13

14

counts = []

t = time.time()

for x in range(10):

    process = Process(target=count, args=(1,1))

    counts.append(process)

    process.start()

e = counts.__len__()

while True:

    for th in counts:

        if not th.is_alive():

            e -= 1

    if e <= 0:

        break

print("Multiprocess cpu", time.time() - t)


Output: 54.342000007629395、53.437999963760376

(10)测试多进程并发执行IO密集型操作


Python

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

t = time.time()

ios = []

t = time.time()

for x in range(10):

    process = Process(target=io)

    ios.append(process)

    process.start()

 

e = ios.__len__()

while True:

    for th in ios:

        if not th.is_alive():

            e -= 1

    if e <= 0:

        break

print("Multiprocess IO", time.time() - t)


Output: 12.509000062942505、13.059000015258789

(11)测试多进程并发执行Http请求密集型操作


Python

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

t = time.time()

httprs = []

t = time.time()

for x in range(10):

    process = Process(target=http_request)

    ios.append(process)

    process.start()

 

e = httprs.__len__()

while True:

    for th in httprs:

        if not th.is_alive():

            e -= 1

    if e <= 0:

        break

print("Multiprocess Http Request", time.time() - t)


Output: 0.5329999923706055、0.4760000705718994

实验结果

 CPU密集型操作IO密集型操作网络请求密集型操作
线性操作94.9182499646922.461999952797.3296000004
多线程操作101.170000076224.86050009730.5053332647
多进程操作53.889999985712.78400003910.5045000315

通过上面的结果,我们可以看到:

  • 多线程在IO密集型的操作下似乎也没有很大的优势(也许IO操作的任务再繁重一些就能体现出优势),在CPU密集型的操作下明显地比单线程线性执行性能更差,但是对于网络请求这种忙等阻塞线程的操作,多线程的优势便非常显著了

  • 多进程无论是在CPU密集型还是IO密集型以及网络请求密集型(经常发生线程阻塞的操作)中,都能体现出性能的优势。不过在类似网络请求密集型的操作上,与多线程相差无几,但却更占用CPU等资源,所以对于这种情况下,我们可以选择多线程来执行


0人推荐
随时随地看视频
慕课网APP