Haven't the time right now to read your post properly, but a comment and a question.
Paste server would likely preload TG. In Apache/mod_wsgi it would lazy load it on demand when first request to a process arrives. How did you ensure your tests against Apache/mod_wsgi weren't counting the cost of loading TG? Graham On May 20, 6:10 am, cd34 <[email protected]> wrote: > I tried multiple options, ending with > > WSGIDaemonProcess daemon processes=8 threads=8 > > where processes was 2x the number of cpus (mutex issues appear to hit > at times) > > number of threads didn't seem to affect performance much after 8. > > I thought the docs section of the turbogears site ran turbogears -- or > is that mod_python/sphinx doing the ajax search? > > In looking at it further, it appears to be an application stack > issue. SQLAlchemy doesn't quite do things in the most efficient > manner when dealing with backreferences, so, I am probably going to > change things around a little to specify the joins myself. > > I guess I need to do a bit of profiling to see where the bottleneck > really is, then, figure out how to engineer around it. > > I'll eventually deploy with nginx, but, the webserver/wsgi end doesn't > appear to be the bottleneck. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "TurboGears" 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/turbogears?hl=en -~----------~----~----~----~------~----~------~--~---
