2009/9/13 Antoni Aloy <antoni.a...@gmail.com>: > 2009/9/13 Bret Taylor <btay...@gmail.com>: >> >> I am one of the authors of Tornado (http://www.tornadoweb.org/), the >> web server/framework we built at FriendFeed that we open sourced last >> week (see http://bret.appspot.com/entry/tornado-web-server). >> >> The underlying non-blocking HTTP server is fairly high performance, so >> I have been working this weekend to get other frameworks like Django >> and web.py working on Tornado's server so existing projects could >> potentially benefit from the performance. To that end, I just checked >> in change to Tornado that enables you to run any WSGI-compatible >> framework on Tornado's HTTP server. You can find it in a class called >> WSGIContainer in our wsgi.py: >> >> http://github.com/facebook/tornado/blob/master/tornado/wsgi.py#L188 >> >> You will have to check out Tornado from github to get the change; it >> is not yet included in the tarball distribution. >> >> Here is a template for running a Django app on Tornado's server using >> the module: >> >> import django.core.handlers.wsgi >> import os >> import tornado.httpserver >> import tornado.ioloop >> import tornado.wsgi >> >> def main(): >> os.environ["DJANGO_SETTINGS_MODULE"] = 'myapp.settings' >> application = django.core.handlers.wsgi.WSGIHandler() >> container = tornado.wsgi.WSGIContainer(application) >> http_server = tornado.httpserver.HTTPServer(container) >> http_server.listen(8888) >> tornado.ioloop.IOLoop.instance().start() >> >> if __name__ == "__main__": >> main() >> >> >> I have only done very basic tests using the new module, so if any of >> you are interested and start using Tornado with your Django projects, >> please let us know what bugs you find so we can fix them. Any and all >> feedback is appreciated. >> > > We're acually using Cherrypy, so it would be quite easy to test and > check the performance. We'll post any issues we found. > > Just few numbers:
ab -c 10 -t 60 http://localhost:8088/ Executing Django with Cherrypy with 3 threads on a PPC Computer 2 CPU 2 GHz With Tornado WSGI: ab -c 10 -t 60 http://localhost:8888/ This is ApacheBench, Version 2.3 <$Revision: 655654 $> Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/ Licensed to The Apache Software Foundation, http://www.apache.org/ Benchmarking localhost (be patient) Completed 5000 requests Completed 10000 requests Completed 15000 requests Completed 20000 requests Completed 25000 requests Finished 25022 requests Server Software: TornadoServer/0.1 Server Hostname: localhost Server Port: 8888 Document Path: / Document Length: 266 bytes Concurrency Level: 10 Time taken for tests: 60.019 seconds Complete requests: 25022 Failed requests: 0 Write errors: 0 Total transferred: 9333206 bytes HTML transferred: 6655852 bytes Requests per second: 416.90 [#/sec] (mean) Time per request: 23.987 [ms] (mean) Time per request: 2.399 [ms] (mean, across all concurrent requests) Transfer rate: 151.86 [Kbytes/sec] received Connection Times (ms) min mean[+/-sd] median max Connect: 0 0 0.1 0 6 Processing: 3 24 1.9 24 68 Waiting: 0 24 1.9 23 68 Total: 7 24 1.9 24 68 Percentage of the requests served within a certain time (ms) 50% 24 66% 24 75% 24 80% 24 90% 25 95% 26 98% 27 99% 28 100% 68 (longest request) Using CherryPy with 3 threads (the best configuration for my computer in my tests) ab -c 10 -t 60 http://localhost:8088/ This is ApacheBench, Version 2.3 <$Revision: 655654 $> Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/ Licensed to The Apache Software Foundation, http://www.apache.org/ Benchmarking localhost (be patient) Completed 5000 requests Completed 10000 requests Completed 15000 requests Completed 20000 requests Finished 21500 requests Server Software: CherryPy/3.0.3 Server Hostname: localhost Server Port: 8088 Document Path: / Document Length: 266 bytes Concurrency Level: 10 Time taken for tests: 60.001 seconds Complete requests: 21500 Failed requests: 0 Write errors: 0 Total transferred: 8299000 bytes HTML transferred: 5719000 bytes Requests per second: 358.33 [#/sec] (mean) Time per request: 27.907 [ms] (mean) Time per request: 2.791 [ms] (mean, across all concurrent requests) Transfer rate: 135.07 [Kbytes/sec] received Connection Times (ms) min mean[+/-sd] median max Connect: 0 0 20.5 0 2999 Processing: 3 28 10.7 26 413 Waiting: 3 26 10.3 24 412 Total: 3 28 23.1 26 3031 Percentage of the requests served within a certain time (ms) 50% 26 66% 28 75% 30 80% 31 90% 34 95% 38 98% 43 99% 48 100% 3031 (longest request) To sumarize: 416.90 req/s for Tornado WSGI 358.33 req/s for CherryPy Tornado makes one CPU up to 100% nearly at the same time of the ab execution, Cherrypy takes longer. Perhaps some CherryPy guru could configure it better. The code is at http://appfusedjango.googlecode.com/svn/trunk/hello_world for the CherryPy (django-cerise) version http://appfusedjango.googlecode.com/svn/trunk/hello_tornado for the tornado version (includes tornado wsgi) This is a sample helloword project with just the essential to run. It's the same code I used to compare with PHP frameworks http://trespams.com/2009/05/10/django-vs-php-framewors/ -- Antoni Aloy López Blog: http://trespams.com Site: http://apsl.net --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Django users" group. To post to this group, send email to django-users@googlegroups.com To unsubscribe from this group, send email to django-users+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/django-users?hl=en -~----------~----~----~----~------~----~------~--~---
