if threads != processes:

nathan at islanddata.com nathan at islanddata.com
Sat Jun 26 20:08:24 CEST 1999

I chose to make my current project multi-threaded rather than forking
for two reasons: speed of context-switches and shared memory.  The
memory part is working out great.  The Queue class seems to be suiting
me just fine.  However, I've noticed that every thread that gets
spawned off is assigned its own process id.  Is this truly less
expensive than forking?  And those processes seem to stick around even
after the target function returns...am I doing something wrong?
Thanks for any input.

More information about the Python-list mailing list