Loading......

Password

What is Multi-tasking?

Definition

Multi-tasking allows your program to perform multiple operations concurrently, improving efficiency when dealing with I/O-bound or external processes.

Three Approaches in Python

# Threading: For I/O-bound tasks (network, disk)
import threading

# Subprocess: For running external programs
import subprocess

# Multiprocessing: For CPU-bound tasks (not covered here)
import multiprocessing

When to Use Which

• Threading: Multiple tasks waiting for responses (web scraping, file downloads)
• Subprocess: Launching external applications (system commands, other programs)

• Multiprocessing: CPU-intensive calculations (data processing, computations)

---

Threading vs Subprocess: Key Differences

Threading

# Threads share the same memory space
shared_list = []

def add_item(item):
    shared_list.append(item)  # All threads access this list

# Memory is shared, communication is easy but risky

Subprocess

# Subprocesses run in isolated memory spaces
# Communication requires explicit channels (pipes, files, stdout)

# No shared memory, completely isolated

Comparison Table

Feature	Threading	Subprocess
Memory	Shared	Isolated
Speed	Fast startup	Slower startup
Use Case	I/O-bound tasks	External programs
Overhead	Low	Higher
Communication	Direct (risky)	Pipes/files

---

Threading

Doing multiple tasks seemingly “at the same time”

# Simple sequential vs concurrent comparison
import time

def task(name, duration):
    print(f"Task {name} started")
    time.sleep(duration)
    print(f"Task {name} completed")

# Sequential execution
task("A", 2)
task("B", 1)

Blocking vs Non-blocking

Blocking:

• Wait until resource becomes available
• Nothing else can happen meanwhile

Non-blocking:

• Try to use resource, move on if busy
• Check back later or get notified

from threading import Thread
import time

def blocking_task():
    print("Blocking task started")
    time.sleep(5)
    print("Blocking task finished")

# Blocking

blocking_task()

this_thread=Thread(target=blocking_task)
this_thread.start()
this_thread.is_alive()
this_thread.join()
this_thread.is_alive()

# Non-blocking

this_thread=Thread(target=blocking_task)
this_thread.start()
print("Main thread continues working")
this_thread.is_alive()

this_thread.is_alive()


## Give arguments
from threading import Thread
import time

def blocking_task(sleepTime=5):
    print("Blocking task started")
    print("Sleep time is:"+str(sleepTime))
    time.sleep(sleepTime)
    print("Blocking task finished")


this_thread=Thread(target=blocking_task,args=(15,))
this_thread.start()
this_thread.is_alive()

this_thread.join()

this_thread.is_alive()



## Get Results
from threading import Thread
import queue
import time

def blocking_task(q, sleepTime=5):
    print("Blocking task started")
    print("Sleep time is:"+str(sleepTime))
    time.sleep(sleepTime)
    print("Blocking task finished")
    result=sleepTime*2
    q.put(result)

q=queue.Queue()
this_thread=Thread(target=blocking_task,args=(q,3))
this_thread.start()
this_thread.is_alive()

this_thread.join()

this_thread.is_alive()

result=q.get()
print(result)

Self-study: “subprocess” library in Python3 (real parallel computing)