[CentralOH] Stackless Python

[Jay Shaffstall]

On Wed, Nov 30, 2011 at 1:27 PM, James -- Atlantix
<james at atlantixeng.com> wrote:
> How to use it is the general question . . . and how many have had good
> results with it? James

It's been a couple of years for me, so I'll refer you to the Stackless
tutorials on how to use it.

What I used it for was a simulation of cavern development in three
dimensions, so the number of nodes that needed processing grew
quickly.  I created coroutines for those nodes that needed processing.

We did have good results, although profiling the app showed that the
initial creation of the coroutines was pretty expensive.  Lazy
creation of the coroutines was necessary to prevent a huge amount of
overhead (in cavern development, many of the nodes would remain static
and never need processed).

We used Stackless specifically so we could process only those nodes
that needed it, and ignore the rest.  Stackless scheduling did that,
in a way that didn't require synchronization headaches.

Jay