Ndtyjky

I decided to create a python code to encrypt my data a while back, when i am half way through, i decided to do a performance test, and the result is horrible, like only 430kB/s encrypting speed. Opening System monitor showed that only 1 thread of my 8C16T processor is being used by my program.

After I try temporary deleting the encrypting part in my code, it was only like 10kB/s faster. So i analyzed the code and found out this part of code

with open("10MB.test", 'rb') as f:  

    byte = f.read(1)

    if(i == 0):

        test = (str("".join([ch.encode("hex")  for line in byte for ch in line])))

        i = 1

        while byte != "":

            g = g + 1

            byte = f.read(1)

            a = str("".join([ch.encode("hex")  for line in byte for ch in line]))

            test = test + a

runs very slow, and system monitor shows that it only uses one thread of my cpu . Is there a way to utilize all thread of my CPU to make this run faster?

import time

from datetime import datetime

realsr = ""

test = ""

i = 0

m = 0

g = 2

j = 0

ti1 = 0

ti2 = 0

ea = raw_input("Please Input Your First Password: ")

print "You Entered:",ea

eb = raw_input("Please Input Your Second Password: ")

print "You Entered:",eb

ec = raw_input("Please Input Your Third Password: ")

print "You Entered:",ec

ed = raw_input("Please Input Your Forth Password: ")

print "You Entered:",ed

ee = raw_input("Please Input Your Fifth Password: ")

print "You Entered:",ee

ef = raw_input("Please Input Your Sixth Password: ")

print "You Entered:",ef

eg = raw_input("Please Input Your Seventh Password: ")

print "You Entered:",eg

eh = raw_input("Please Input Your Eighth Password: ")

print "You Entered:",eh

ti1 = int(round(time.time() * 1000))

with open("10MB.test", 'rb') as d:

    j = len(str("".join([ch.encode("hex")  for line in d for ch in line])))

with open("10MB.test", 'rb') as f:  

byte = f.read(1)

if(i == 0):

    test = (str("".join([ch.encode("hex")  for line in byte for ch in line])))

    i = 1

while byte != "":

    g = g + 1

    byte = f.read(1)

    a = str("".join([ch.encode("hex")  for line in byte for ch in line]))

    test = test + a

    if(j != len(test)):

        temp1 = int(a,16)

        if(m == 0):

            temp1 * ea

        if(m == 1):

            temp1 * eb

        if(m == 2):

            temp1 * ec

        if(m == 3):

            temp1 * ed

        if(m == 4):

            temp1 * ee

        if(m == 5):

            temp1 * ef

        if(m == 6):

            temp1 * eg

        if(m == 7):

            temp1 * eh

while len(test) >= j:

    with open("clone.test", 'wb') as k:

        k.write(test.decode("hex"))

    ti2 = int(round(time.time() * 1000))

    print "done!"

    print test

    print (ti2-ti1)

    print ti1

    print (ti2)

    break

You read your file one byte at a time: f.read(1). This is a horrible idea. You should read as much as you can safely fit in the memory (perhaps the whole file, if the memory size permits) and then do encryption.

If you need to access all your CPU cores you will need use Pythons Multiprocessing module.

And from the docs;

In multiprocessing, processes are spawned by creating a Process object
and then calling its start() method. Process follows the API of
threading.Thread. A trivial example of a multiprocess program is

from multiprocessing import Process



def f(name):

    print('hello', name)



if __name__ == '__main__':

    p = Process(target=f, args=('bob',))

    p.start()

    p.join()

Instead of using "byte = f.read(1)" in order to load 1 byte from the file, use this :

content = f.read()

for byte in content:

    # your code

With this method, you could won in speed execution because you don't stop reading between all bytes. You only access the file once.