Hi Mars,
  One possibly important difference is that items in app.yaml are
separate WSGI apps.  So if you have distinct sections of your
application, such as backend services and frontend views, that don't
share a lot of code between them, a loading request will only need to
load the modules used by the WSGI app the url that got hit maps to in
app.yaml.  This could have an impact if you're loading a lot of
unneeded modules, or you if you've got some very rarely used modules.
This difference is probably less important now with warming requests;
recently there seem to be far fewer issues with corrupted instances on
spinup.

  Personally, if I have something that is logically a separate
component that is very 'modular' I often define it in app.yaml.  Also,
I often define components that are relatively infrequently used (like
dev / admin stuff) as a separate apps, since there is not much point
in loading that stuff most of the time.



Robert



On Mon, May 2, 2011 at 23:37, Mars <[email protected]> wrote:
> If I have a large number of url patterns to match, is it more
> efficient, in terms of performance, to do it in app.yaml or passing
> them as arguments to WSGIApplication constructor?
>
> p.s. I'm using Python, but I'd imagine similar question applies to
> Java?
>
> Cheers,
>
> Mars
>
> --
> You received this message because you are subscribed to the Google Groups 
> "Google App Engine" group.
> To post to this group, send email to [email protected].
> To unsubscribe from this group, send email to 
> [email protected].
> For more options, visit this group at 
> http://groups.google.com/group/google-appengine?hl=en.
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/google-appengine?hl=en.

Reply via email to