threads_vs_process.txt

# Both are ways to achieve multi tasking.
# threads are created by a process inside process.
# threads share same address space.
# processes has their own adress space.
# data between process can't be accessed without special inter process communication.
# like files, shared  memory, message pipe.
#  multiple processes can't share memory. they are isolated.

# process: data, files, registers, stack, code/codes
# thread: code is executed by thread. register and stack are different for threads.

# threads is same like asyncronous programming in python. it's not true parallelism.
# while process is true parallelism, it can run on different cpu cores.

# GIL locks is python mutex/lock that allows only one thread to get hold of interpreter.
# this is because python uses reference counting for memory management.
# objects created in python have reference count which keeps track of number of references poiting to object.
# when this count reaches zero, object is removed and memory is released.
# 2 threads can increase/decrease the value of reference count simultaneously which might cause issue.

# In multiprocessing, it creates multiple interpreter in python. thus it can run simultaneously.