[CentralOH] Fwd: Re: Stackless Python

William McVey wam at cisco.com
Thu Oct 29 15:09:01 CET 2009


On Wed, 2009-10-28 at 17:30 -0400, Bryan Harris wrote:
> The processing module sticks each thread in a different python
> process. The syntax is identical to the threading library so it's two
> lines of code to switch from threating to processing. I'm going to try
> that and see how much better it does.

As of python 2.6, the third party library 'processing' (aka
pyprocessing) was extended and incorporated into the python standard lib
as the 'multiprocessing' module. PEP-371
(http://www.python.org/dev/peps/pep-0371/) has details of the module
becoming part of the standard libary. After the 2.6, the updated
'multiprocessing' library was backported to older versions of python
(2.4 and 2.5) and is available at:
http://pypi.python.org/pypi/multiprocessing/

Unless you're supporting legacy code that already was written with the
old processing/pyprocessing library, I don't see any reason to use it
rather than multiprocessing.

  -- William

P.S. With all that being said, multiprocessing is a *great* library for
managing concurrency. I've ported most of stuff that used to use the
threading library to using it and I've been very happy. About the only
thing I use threads for nowdays is for client side GUI interfaces (which
thanks to HTML/AJAX are becoming few and far between, for me at least).



More information about the CentralOH mailing list