LoginSignup
0
0

More than 1 year has passed since last update.

Memo - Python Parallel and Concurrent Programming

Last updated at Posted at 2022-08-12

Threading

to run the function in a new thread (non-daemon)

import threading
thread_variable = threading.Thread(target=function_name)
thread_variable.start()

to call the function back to the main thread

thread_variable.join()

to run the function in a daemon thread

When the main thread exits, the daemon thread(s) will automatically exits.

thread_variable.daemon = True
thread_variable.start()

to set a mutual exclusion

lock_variable = threading.Lock()

to protect "my" access

Will wait if someone else is accessing

lock_variable.acquire()

to protect "my" access without waiting

Will give up immediately if not available

lock_variable.acquire(blocking=False)

to release "my" access

lock_variable.release()

to use Reentrant lock

This allows to lock multiple times in the same thread. It has to unlock as many as locking. Useful in a recurrsion.

rlock_variable = threading.RLock()
rlock_variable.acquire()
rlock_variable.release()

to lock conditionally

lock_condition_variable = threading.Condition(lock=a_variable)
to wait until being notified by other threads
lock_condition_variable.wait()
to notify another thread

Should call before exiting a thread

lock_condition_variable.notify()
to notify all other threads before exiting a thread

Should call before exiting a thread

lock_condition_variable.notify_all()

to use Read/Write lock

from readerwriterlock import rwlock
rwlock_variable = rwlock.RWLockFair()

to lock for reading

read_lock_variable = rwlock_variable.gen_rlock()
read_lock_variable.acquire()
read_lock_variable.release()

to lock for writing

write_lock_variable = rwlock_variable.gen_wlock()
write_lock_variable.acquire()
write_lock_variable.release()

to use Semaphore

Semaphore manages a counter representing the number of release() calls minus the number of acquire() calls.

lock_variable = threading.Semaphore(1)

Thread Pool

Maintain a collection of threads and re-use them.

from concurrent.futures import ThreadPoolExecutor
pool = ThreadPoolExecutor(max_workers=5)
pool.submit(function_name)
pool.shutdown()

Future

from concurrent.futures import ThreadPoolExecutor
pool = ThreadPoolExecutor(max_workers=5)
future = pool.submit(function_name)
future.result()

Multiprocessing

to start the function in a new process

import multiprocessing
process_variable = multiprocessing.Process(target=function_name)
process_variable.start()

Process Pool

Main a collection of threads and re-use them.

from concurrent.futures import ProcessPoolExecutor
pool = ProcessPoolExecutor(max_workers=5)
pool.submit(function_name)
pool.shutdown()
0
0
1

Register as a new user and use Qiita more conveniently

  1. You get articles that match your needs
  2. You can efficiently read back useful information
  3. You can use dark theme
What you can do with signing up
0
0