Re: nasa site on django

2007-07-04 Thread Graham Dumpleton
On Jul 4, 11:20 pm, Gábor Farkas <[EMAIL PROTECTED]> wrote: > Jeremy Dunck wrote: > >> Frankly I think a generic python application server similar to tomcat > >> would do a world of good for python apps in general. Something similar to > >> cherrypy... but this is beyond the responsibilities of

Re: nasa site on django

2007-07-04 Thread Gábor Farkas
Jeremy Dunck wrote: >> Frankly I think a generic python application server similar to tomcat >> would do a world of good for python apps in general. Something similar to >> cherrypy... but this is beyond the responsibilities of the django community >> which have plenty great work left. > > Simi

Re: nasa site on django

2007-06-28 Thread Jacob Kaplan-Moss
On 6/28/07, Vance Dubberly <[EMAIL PROTECTED]> wrote: > Django suffers from running under mod_python in > a prefork environment where every process loads a full copy of django. So > as I'm sure you can imagine RAM gets chewed up very quickly. Are you running Django behind a load balancer like Pe

Re: nasa site on django

2007-06-28 Thread Jeremy Dunck
On 6/28/07, Vance Dubberly <[EMAIL PROTECTED]> wrote: > However resource intensity is a very big deal. RAM is a huge factor > here. Thanks for clearing up the issue. It's true that the process-per-request model is heavier in terms of RAM than alternatives, but it's generally fast. Maybe your

Re: nasa site on django

2007-06-28 Thread Vance Dubberly
Jeremy, I don't know about Java vs Python speeds. I've little doubt Java is alot faster if for no other reason than the amount of resources Sun puts behind it. However resource intensity is a very big deal. RAM is a huge factor here. The Tomcat App was running in a VM limited 256 meg and ser

Re: nasa site on django

2007-06-28 Thread Vance Dubberly
No dynamically assembling the pages via django is faster than serving static files off disk with Apache. Vance On 6/27/07, KpoH <[EMAIL PROTECTED]> wrote: > > > I don't understand. Static files served by django? > > Vance Dubberly wrote: > > Moved http://opensource.arc.nasa.gov to django a couple

Re: nasa site on django

2007-06-28 Thread Vance Dubberly
Kai, Very very untrue. RAM is much faster than Disk IO. Dynamically assembling a page from query results cached in database memory is far faster than going to disk. Caching the rendered page via memcached, squid, or even in a database table is so much faster than serving static disk bound fi

Re: nasa site on django

2007-06-28 Thread Jacob Kaplan-Moss
On 6/27/07, Vance Dubberly <[EMAIL PROTECTED]> wrote: > Moved http://opensource.arc.nasa.gov to django a couple of weeks ago. > Thought ya'll might want to know that. Congrats :) > Also thought you might want to know the site was running on Tomcat/Mysql and > the performance difference is mind b

Re: nasa site on django

2007-06-27 Thread Jeremy Dunck
On 6/27/07, Vance Dubberly <[EMAIL PROTECTED]> wrote: ... > ball of red tape but let's just say django is ALOT more resource intensive > and ALOT slower. Have you done any profiling? I haven't compared such a port, but I'd be surprised if performance is significantly (i.e. one order) different w

Re: nasa site on django

2007-06-27 Thread Kai Kuehne
Hi, > Vance Dubberly wrote: > > On the plus side it is faster than a similar PHP app ( with APC ) and > > faster than serving static files. Afaik, dynamic things _can't be_ faster than static things. Kai --~--~-~--~~~---~--~~ You received this message because yo

Re: nasa site on django

2007-06-27 Thread KpoH
I don't understand. Static files served by django? Vance Dubberly wrote: > Moved http://opensource.arc.nasa.gov to django a couple of weeks ago. > Thought ya'll might want to know that. > > Also thought you might want to know the site was running on > Tomcat/Mysql and the performance difference