Pythonでマルチスレッドとマルチプロセス

### シングルスレッド

from concurrent.futures import ThreadPoolExecutor
import time
def func():
    time.sleep(1)
start = time.time()
for i in range(8):
    func()
print(time.time() - start)

$ python3 single.py
8.064586639404297

### マルチスレッド(同時に実行)

from concurrent.futures import ThreadPoolExecutor
import time
def func():
    time.sleep(1)
start = time.time()
with ThreadPoolExecutor(max_workers=4) as e:
    for i in range(8):
        e.submit(func)
print(time.time() - start)

$ python3 multithread.py
2.0498673915863037

### マルチプロセス(複数のプロセスで同時に処理)

from concurrent.futures import ProcessPoolExecutor
import time
def func():
    time.sleep(1)
start = time.time()
with ProcessPoolExecutor(max_workers=4) as e:
    for i in range(8):
        e.submit(func)
print(time.time() - start)

$ python3 multiprocess.py
2.1424620151519775

import os
import time
import datetime
from concurrent.futures import ProcessPoolExecutor, ThreadPoolExecutor

x1, x2, y1, y2 = -1.8, 1.8, -1.8, 1.8
c_real, c_imag = -0.62772, -0.42193

def calculate_z_serial_purepython(maxiter, zs, cs):
    output = [0] * len(zs)
    for i in range(len(zs)):
        n = 0
        z = zs[i]
        c = cs[i]
        if (i % 100) == 0:
            time.sleep(0.0001)
        while abs(z) < 2 and n < maxiter:
            z = z * z + c
            n += 1
        output[i] = n
    return output

def calc_pure_python(desired_width, max_iterations):
    x_step = (float(x2 - x1) / float(desired_width))
    y_step = (float(y1 - y2) / float(desired_width))
    x = []
    y = []
    ycoord = y2
    while ycoord > y1:
        y.append(ycoord)
        ycoord += y_step
    xcoord = x1
    while xcoord < x2:
        x.append(xcoord)
        xcoord += x_step

    zs = []
    cs = []
    for ycoord in y:
        for xcoord in x:
            zs.append(complex(xcoord, ycoord))
            cs.append(complex(c_real, c_imag))
    
    output = calculate_z_serial_purepython(max_iterations, zs, cs)

if __name__ == "__main__":

    max_workers = os.cpu_count()
    start = datetime.datetime.now()
    for i in range(16):
        calc_pure_python(desired_width=500, max_iterations=100)
    elapsed = datetime.datetime.now() - start
    print("SingleThread: {}ms".format(elapsed.seconds*1000 + elapsed.microseconds/1000))

    start = datetime.datetime.now()
    with ThreadPoolExecutor(max_workers=max_workers) as executor:
        for i in range(16):
            executor.submit(calc_pure_python, 500, 100)
    elapsed = datetime.datetime.now() - start
    print("MultiThread: {}ms".format(elapsed.seconds*1000 + elapsed.microseconds/1000))

    # マルチプロセスの場合
    start = datetime.datetime.now()
    with ProcessPoolExecutor(max_workers=max_workers) as executor:
        for i in range(16):
            executor.submit(calc_pure_python, 500, 100)
    elapsed = datetime.datetime.now() - start
    print("MultiProcess: {}ms".format(elapsed.seconds*1000 + elapsed.microseconds/1000))

$ python3 main.py
SingleThread: 17934.699ms
MultiThread: 11256.051ms
MultiProcess: 8493.925ms

os.cpu_count()　でCPUのコアを取得できるんか…
処理に時間がかかる箇所をマルチプロセスで実装すれば良さそうではあるな…