Yes. eventlet failed the test. OSX 10.6.3 and python 2.6
On Dec 5, 1:34 am, Phyo Arkar <phyo.arkarl...@gmail.com> wrote: > This discussion becomes not about comparison of scalability of web2py > or scability of web2py vs other framework but web2py vs other > langauges?? > > >massimo wrote: > >ab -n 10000 -c 100http://127.0.0.1:8081/ > > >rocket: 0.629 [ms] > >eventlet: > > Massimo , does eventlet failed in that test? > > On 12/5/10, mdipierro <mdipie...@cs.depaul.edu> wrote: > > > This is supposed to be even better > > > #http://code.google.com/p/gevent/source/browse/examples/wsgiserver.py > > from gevent import wsgi > > > but cannot install on snow leopard. > > > Massimo > > > On Dec 4, 9:19 pm, mdipierro <mdipie...@cs.depaul.edu> wrote: > >> I tried this: > > >> #http://eventlet.net/doc/examples.html#wsgi-server > >> import eventlet > >> from eventlet import wsgi > >> from rocket import Rocket > > >> def hello_world(env, start_response): > >> if env['PATH_INFO'] != '/': > >> start_response('404 Not Found', [('Content-Type', 'text/ > >> plain')]) > >> return ['Not Found\r\n'] > >> start_response('200 OK', [('Content-Type', 'text/plain')]) > >> return ['Hello, World!\r\n'] > > >> if __name__=='__main__': > >> if True: > >> r=Rocket(('127.0.0.1',8081),'wsgi', {'wsgi_app':hello_world}) > >> r.start() > >> else: > >> wsgi.server(eventlet.listen(('127.0.0.1', 8081)), hello_world) > > >> with > > >> ab -n 10000 -c 10http://127.0.0.1:8081/ > > >> rocket: 0.618 [ms] > >> eventlet: 0.443 [ms] > > >> ab -n 10000 -c 100http://127.0.0.1:8081/ > > >> rocket: 0.629 [ms] > >> eventlet: > > >> Benchmarking 127.0.0.1 (be patient) > >> Completed 1000 requests > >> Test aborted after 10 failures > >> apr_socket_connect(): Connection reset by peer (54) > >> Total of 1998 requests completed > > >> On Dec 4, 7:39 pm, Branko Vukelic <bg.bra...@gmail.com> wrote: > > >> > On Sun, Dec 5, 2010 at 1:59 AM, blackthorne <francisco....@gmail.com> > >> > wrote: > >> > > I've read it while ago. > >> > > The "problem" with that test is the number of processors. It takes a > >> > > high number of CPUs to bring Erlang benefits. > > >> > Another 'problem' is that it's not about performance when it comes to > >> > Erlang. It's about overall robustness. For example, Yaws HTTP server > >> > may not be the fastest around, but you just cannot kill it. Even if it > >> > drops a request, it will keep on running, and handling whatever > >> > requests you throw at it. I guess I had that in mind when I said > >> > scalability. > > >> > Also, Erlang has software threads, afaik, not hardware CPU threads, > >> > and it manages those internally using a supervisor-worker > >> > architecture. That's something built into the language, and you mostly > >> > don't have to worry about it. > > >> > -- > >> > Branko Vukelić > > >> > bg.bra...@gmail.com > >> > stu...@brankovukelic.com > > >> > Check out my blog:http://www.brankovukelic.com/ > >> > Check out my portfolio:http://www.flickr.com/photos/foxbunny/ > >> > Registered Linux user #438078 (http://counter.li.org/) > >> > I hang out on identi.ca:http://identi.ca/foxbunny > > >> > Gimp Brushmakers Guildhttp://bit.ly/gbg-group > >
