[Python-Dev] Yet another "A better story for multi-core Python" comment

[Stephen J. Turnbull]

R. David Murray writes:
 > On Tue, 08 Sep 2015 10:12:37 -0400, Gary Robinson <garyrob at me.com> wrote:
 > > 2) Have a mode where a particular data structure is not reference
 > > counted or garbage collected.
 > 
 > This sounds kind of like what Trent did in PyParallel (in a more generic
 > way).

Except Gary has a large persistent data structure, and Trent's only
rule is "don't persist objects you want to operate on in parallel."
The similarity may be purely superficial, though.

@Gary: Justifying your request is unnecessary.  As far as I can see,
everybody acknowledges that "large shared data structure" + "multiple
cores" is something that Python doesn't do well enough in some sense.
It's just a hard problem, and the applications that really need it are
sufficiently specialized that we haven't been able to justify turning
the world upside down to serve them.

Trent seems to be on to something that requires only a bit of a tilt
;-), and despite the caveat above, I agree with David, check it out:

https://mail.python.org/pipermail/python-dev/2015-September/141485.html