Closes #21527: Add default number of workers to ThreadPoolExecutor. (Claudiu Popa.)

4 files changed, 24 insertions, 2 deletions

diff --git a/Doc/library/concurrent.futures.rst b/Doc/library/concurrent.futures.rst
index 08c926a..e487817 100644
--- a/Doc/library/concurrent.futures.rst
+++ b/Doc/library/concurrent.futures.rst
@@ -115,11 +115,19 @@ And::
    executor.submit(wait_on_future)
 
 
-.. class:: ThreadPoolExecutor(max_workers)
+.. class:: ThreadPoolExecutor(max_workers=None)
 
    An :class:`Executor` subclass that uses a pool of at most *max_workers*
    threads to execute calls asynchronously.
 
+   .. versionchanged:: 3.5
+      If *max_workers* is ``None`` or
+      not given, it will default to the number of processors on the machine,
+      multiplied by ``5``, assuming that :class:`ThreadPoolExecutor` is often
+      used to overlap I/O instead of CPU work and the number of workers
+      should be higher than the number of workers
+      for :class:`ProcessPoolExecutor`.
+
 
 .. _threadpoolexecutor-example:
 
diff --git a/Lib/concurrent/futures/thread.py b/Lib/concurrent/futures/thread.py
index 8d6081c..3ae442d 100644
--- a/Lib/concurrent/futures/thread.py
+++ b/Lib/concurrent/futures/thread.py
@@ -10,6 +10,7 @@ from concurrent.futures import _base
 import queue
 import threading
 import weakref
+import os
 
 # Workers are created as daemon threads. This is done to allow the interpreter
 # to exit when there are still idle threads in a ThreadPoolExecutor's thread
@@ -80,13 +81,17 @@ def _worker(executor_reference, work_queue):
         _base.LOGGER.critical('Exception in worker', exc_info=True)
 
 class ThreadPoolExecutor(_base.Executor):
-    def __init__(self, max_workers):
+    def __init__(self, max_workers=None):
         """Initializes a new ThreadPoolExecutor instance.
 
         Args:
             max_workers: The maximum number of threads that can be used to
                 execute the given calls.
         """
+        if max_workers is None:
+            # Use this number because ThreadPoolExecutor is often
+            # used to overlap I/O instead of CPU work.
+            max_workers = (os.cpu_count() or 1) * 5
         if max_workers <= 0:
             raise ValueError("max_workers must be greater than 0")
 
diff --git a/Lib/test/test_concurrent_futures.py b/Lib/test/test_concurrent_futures.py
index 83184c3..11560e6 100644
--- a/Lib/test/test_concurrent_futures.py
+++ b/Lib/test/test_concurrent_futures.py
@@ -11,6 +11,7 @@ test.support.import_module('threading')
 
 from test.script_helper import assert_python_ok
 
+import os
 import sys
 import threading
 import time
@@ -444,6 +445,11 @@ class ThreadPoolExecutorTest(ThreadPoolMixin, ExecutorTest, unittest.TestCase):
         self.executor.shutdown(wait=True)
         self.assertCountEqual(finished, range(10))
 
+    def test_default_workers(self):
+        executor = self.executor_type()
+        self.assertEqual(executor._max_workers,
+                         (os.cpu_count() or 1) * 5)
+
 
 class ProcessPoolExecutorTest(ProcessPoolMixin, ExecutorTest, unittest.TestCase):
     def test_killed_child(self):
diff --git a/Misc/NEWS b/Misc/NEWS
index 71ab268..a02e8a6 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -129,6 +129,9 @@ Core and Builtins
 Library
 -------
 
+- Issue #21527: Add a default number of workers to ThreadPoolExecutor equal
+  to 5 times the number of CPUs.  Patch by Claudiu Popa.
+
 - Issue #22216: smtplib now resets its state more completely after a quit.  The
   most obvious consequence of the previous behavior was a STARTTLS failure
   during a connect/starttls/quit/connect/starttls sequence.