Pourquoi un programme multithreading est-il plus lent qu'un programme à thread unique, alors qu'il lit des fichiers txt séparés?

J'ai plusieurs fichiers txt qui contiennent des informations CDR, les CDR sont distribués dans plusieurs fichiers. J'ai besoin de trouver des téléphones dans ces fichiers, puis de vérifier les correspondances avec le fichier xls. J'ai écrit une version à thread unique, puis multithread et j'ai trouvé que parfois le multithread est plus lent que le thread unique.

Multithread:

import re
import os
import time
import sys
import pandas
import ipaddress


def improve_view_n(string_to_improve):
    string_to_improve = string_to_improve.split(',')
    improved_string = ""
    for i in string_to_improve:
        if i != "":
            improved_string = improved_string + i + "  "
    return improved_string


def upload_number_list():
    numbers = []
    try:
        with open(file="number_list.txt", mode="r") as f:
            for i in f:
                numbers.append(i.strip("\\\n"))
    except FileNotFoundError:
        print("number_list.txt file does not exist or corrupted.\n\n")
        print("The program will be terminated in 5 seconds")
        time.sleep(5)
        sys.exit()
    return numbers


def search_for_pattern(number):
    found_ip = []
    our_files = ('y.py', "found_results.txt", "number_list.txt", 'norma.xls')
    list_files = os.listdir()
    for file_name in list_files:
        #if file_name not in our_files:
        if file_name.startswith("MSK"):
            with open(file=file_name, mode='r') as f:
                text_of_file = f.read()
                results = re.findall(pattern=f",,,,,.*{number}.*,", string=text_of_file)
                if results:
                    for element in results:
                        write_searh_results_to_file(file_name, element)
                        element = improve_view_n(element).split()
                        for subeleement in element:
                            try:
                                ipaddress.IPv4Address(subeleement)
                            except ipaddress.AddressValueError:
                                pass
                            else:
                                found_ip.append(subeleement)
                else:
                    nothing_was_found(file_name, number)
    return found_ip


def write_searh_results_to_file(file_where_match_was_found, element):
    with open(file="found_results.txt", mode='a') as f:
        f.write(f"{file_where_match_was_found}: {improve_view_n(element)} \n")


def nothing_was_found(file_where_match_wasnt_found, number_to_search):
    with open(file="found_results.txt", mode='a') as f:
        f.write(f"NO MATCHES FOUND FOR {number_to_search} IN {file_where_match_wasnt_found}\n\n")


def check_if_ip_in_norma(ip, trunk_names):
    line_which_contains_ip = []
    for line in trunk_names:
        if ip in line:
            line_which_contains_ip.append(line)
    if line_which_contains_ip == []:
        line_which_contains_ip.append(f"Norma does not contain information about {ip}")
    return line_which_contains_ip


def main():
    t1 = int(round(time.time() * 1000))
    found_ip_lists = []
    found_ip_list = []
    if "norma.xls" not in os.listdir():
        print("norma.xls file was not found in the current directory")
        print("The program will be terminated")
        sys.exit()
        time.sleep(3)
    normafile = pandas.read_excel('norma.xls', skiprows=2, header=None)
    trunk_names = normafile[2]
    numbers_to_search_list = upload_number_list()
    for i in numbers_to_search_list:
        found_ip_lists.append(search_for_pattern(i))
    for i in found_ip_lists:
        found_ip_list += i
    print(set(found_ip_list))
    for ip in set(found_ip_list):
        x = check_if_ip_in_norma(ip, trunk_names)
        print(f"{x}\n")
        with open('found_results.txt', 'a') as f:
            f.write(f"{x}\n")
    print("The program completed fine!")
    print("Take found_results.txt from the current folder")
    print("If you want to repeat search, remove found_results.txt")
    t2 = int(round(time.time() * 1000))
    print(f"Job is done within {t2 - t1} miliseconds")
    time.sleep(90)
    print("Bye!")
    time.sleep(1)


if __name__ == '__main__':
    try:
        main()
    except Exception as ex:
        print("The following error happened:")
        print(ex)
        time.sleep(20)

Single Thread:

import re
import os
import time
import sys
import pandas
import ipaddress
import threading


def improve_view_n(string_to_improve):
    string_to_improve = string_to_improve.split(',')
    improved_string = ""
    for i in string_to_improve:
        if i != "":
            improved_string = improved_string + i + "  "
    return improved_string


def upload_number_list():
    numbers = []
    try:
        with open(file="number_list.txt", mode="r") as f:
            for i in f:
                numbers.append(i.strip("\\\n"))
    except FileNotFoundError:
        print("number_list.txt file does not exist or corrupted.\n\n")
        print("The program will be terminated in 5 seconds")
        time.sleep(5)
        sys.exit()
    return numbers


def search_for_pattern(number, file_name, semaphore, found_ip):
    semaphore.acquire()
    if file_name.startswith("MSK"):
        with open(file=file_name, mode='r') as f:
            text_of_file = f.read()
            results = re.findall(pattern=f",,,,,.*{number}.*,", string=text_of_file)
            if results:
                for element in results:
                    write_searh_results_to_file(file_name, element)
                    element = improve_view_n(element).split()
                    for subeleement in element:
                        try:
                            ipaddress.IPv4Address(subeleement)
                        except ipaddress.AddressValueError:
                            pass
                        else:
                            found_ip.append(subeleement)
            else:
                nothing_was_found(file_name, number)
    semaphore.release()


def write_searh_results_to_file(file_where_match_was_found, element):
    with open(file="found_results.txt", mode='a') as f:
        f.write(f"{file_where_match_was_found}: {improve_view_n(element)} \n")


def nothing_was_found(file_where_match_wasnt_found, number_to_search):
    with open(file="found_results.txt", mode='a') as f:
        f.write(f"NO MATCHES FOUND FOR {number_to_search} IN {file_where_match_wasnt_found}\n\n")


def check_if_ip_in_norma(ip, trunk_names):
    line_which_contains_ip = []
    for line in trunk_names:
        if ip in line:
            line_which_contains_ip.append(line)
    if line_which_contains_ip == []:
        line_which_contains_ip.append(f"Norma does not contain information about {ip}")
    return line_which_contains_ip


def main():
    threads = []
    our_files = ('y.py', "found_results.txt", "number_list.txt", 'norma.xls', 'MultyThread.py')
    list_files = os.listdir()
    for file in our_files:
        if file in list_files:
            list_files.remove(file)
    semaphore = threading.Semaphore(10)
    t1 = int(round(time.time() * 1000))
    found_ip_list = []
    if "norma.xls" not in os.listdir():
        print("norma.xls file was not found in the current directory")
        print("The program will be terminated")
        sys.exit()
        time.sleep(3)
    normafile = pandas.read_excel('norma.xls', skiprows=2, header=None)
    trunk_names = normafile[2]
    numbers_to_search_list = upload_number_list()
    for number in numbers_to_search_list:
        for file_number in range(len(list_files)):
            threads.append(threading.Thread(target=search_for_pattern,
                                            args=(number, list_files[file_number],
                                                  semaphore, found_ip_list,),)
                           )
            threads[file_number].start()
        for file_number in range(len(list_files)):
            threads[file_number].join()
    print(set(found_ip_list))
    for ip in set(found_ip_list):
        x = check_if_ip_in_norma(ip, trunk_names)
        print(f"{x}\n")
        with open('found_results.txt', 'a') as f:
            f.write(f"{x}\n")
    print("The program completed fine!")
    print("Take found_results.txt from the current folder")
    print("If you want to repeat search, remove found_results.txt")
    t2 = int(round(time.time() * 1000))
    print(f"Job is done within {t2 - t1} miliseconds")
    time.sleep(90)
    print("Bye!")
    time.sleep(1)


if __name__ == '__main__':
    try:
        main()
    except Exception as ex:
        print("The following error happened:")
        print(ex)
    time.sleep(20)