對比實驗
資料顯示,如果多執行緒的進程是CPU密集型的,那多執行緒並不能有多少效率上的提升,相反還可能會因為執行緒的頻繁切換,導致效率下降,推薦使用多進程;如果是IO密集型,多執行緒程序可以利用IO阻塞等待時的空閒時間執行其他執行緒,提升效率。所以我們根據實驗比較不同場景的效率
(1)引入所需要的模組
import requests import time from threading import Thread from multiprocessing import Process
(2)定義CPU密集的計算函數
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)定義IO密集的文件讀寫函數
reee(4) 定義網路請求函數
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()
(5)測試線性執行IO密集操作、CPU密集操作所需時間、網路請求密集操作所需時間
_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}
輸出
CPU密集操作所需時間、網路請求密集操作所需時間
# 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)
輸出
CPU密集操作:95.6059999466、91.57039862095386395095. 99.96799993515015
IO密集:24.25、21.76699995994568、21.76999980926514、22.060999876999980926514、22.060999870300293319893896. 563999891281128、4.371000051498413、4.522000074386597、14.67100003814697
(6)測試多執行緒並發執行密集作業所需的時間99.9240000248 、101.26400017738342、102.32200002670288
(7)測試多執行緒並發執行IO密集操作所需時間
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)
put: 測試多執行緒並發執行IO密集作業所需時間
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)
68
(8)測試多執行緒並發執行網路密集操作所需時間
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)測試多進程並發執行CPU密集操作所需時間
rr 963760376(10)測試多進程並發執行IO密集型操作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)
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)
透過上面的結果,我們可以看到:
多執行緒在IO密集型的操作下似乎也沒有很大的優勢(也許IO操作的任務再繁重一些就能體現出優勢),在CPU密集型的操作下明顯地比單線程線性執行性能更差,但是對於網絡請求這種忙等阻塞執行緒的操作,多執行緒的優勢便非常顯著了
多進程無論是在CPU密集型或IO密集型以及網路請求密集型(經常發生執行緒阻塞的操作)中,都能體現出效能的優勢。不過在類似網路請求密集的操作上,與多執行緒相差無幾,但卻更佔用CPU等資源,所以對於這種情況下,我們可以選擇多執行緒來執行