Hey, I have just installed apache + web2py + mod_wsgi. Now results are very comparable to django results. Great !
I'm going to test it a bit more. Thanks Daniel On Sun, Jun 21, 2009 at 2:01 AM, AchipA <attila.cs...@gmail.com> wrote: > > You didn't say what web server you used for Django, whether you > benchmarked compiled applications or not, etc, etc. The only way it's > not apples and oranges if you use the SAME web server and compiled > code. You wouldn't be using the bundled servers in a production > environment anyway (that's true for both Web2py and Django). So, set > them both up with, say, Apache and 2 mod_wsgi processes, deploy > compiled test apps and THEN you can start to compare results. > > PS for the record, I'm handling 40-50 requests per sec with a 'real' > web2py (deployment optimized) site on a small Amazon EC2 instance, > which is considerably lower specced then your test machine. > > On Jun 20, 11:23 pm, Daniel Guryca <dun...@gmail.com> wrote: > > Nope I'm not comparing apples with oranges. > > Numbers I gave you are numbers with disabled database stuff (commented > out > > everything in db.py) + super simple Hello world page. > > Even If I enabled database numbers should be much higher - or at least I > > would expect much higher numbers. > > > > I know that python is not java but Django numbers are far better. > > > > regards > > Daniel > > > > On Sat, Jun 20, 2009 at 4:57 PM, mdipierro <mdipie...@cs.depaul.edu> > wrote: > > > > > You are comparing apples with oranges. By bechmarking /examples/ > > > default/index you are not just benchmarking the cherrypy wsgiserver. > > > You have a lot of overhead (sessions, database IO, complex template > > > parsing, file IO, etc.). > > > > > There are different issues here: > > > > > - Cherrypy wsgiserver is probably as fast as Apache+mod_wsgi although > > > not recommended in production (because of all the extra features of > > > apache). Alache should be faster at serving static files because it > > > will bypass some of the web2py logic. > > > - You should make a simpler app and benchmark that > > > - There are many possible optimizations in web2py that I have > > > described in previous posts in this thread. > > > > > Massimo > > > > > On Jun 20, 8:21 am, carlo <syseng...@gmail.com> wrote: > > > > Anyway I would be surprised to see CherryPy+SQLite performing slower > > > > than Apache+anything else at least under moderate load. > > > > > > carlo > > > > > > On 19 Giu, 20:18, Fran <francisb...@googlemail.com> wrote: > > > > > > > On 19 June, 16:01, Daniel Guryca <dun...@gmail.com> wrote: > > > > > > > > I can see that performance coming from a default integrated > server is > > > > > > somewhat poor. > > > > > > > 'integrated server' = CherryPy/SQLite, right? > > > > > > > > What other deployment possibilities could I test ? > > > > > > > ApacheWSGI/PostgreSQL would be a better comparator. > > > > > > > F > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "web2py Web Framework" group. To post to this group, send email to web2py@googlegroups.com To unsubscribe from this group, send email to web2py+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/web2py?hl=en -~----------~----~----~----~------~----~------~--~---
