Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #29800 > unrolled thread

Fastest web framework

Back to article view | Back to comp.lang.python

Contents

  Fastest web framework Andriy Kornatskyy <andriy.kornatskyy@live.com> - 2012-09-23 12:19 +0300
    Re: Fastest web framework Roy Smith <roy@panix.com> - 2012-09-23 10:02 -0400
      Re: Fastest web framework Stefan Behnel <stefan_ml@behnel.de> - 2012-09-23 17:50 +0200
      Re: Fastest web framework Mark Lawrence <breamoreboy@yahoo.co.uk> - 2012-09-23 18:20 +0100
      RE: Fastest web framework Andriy Kornatskyy <andriy.kornatskyy@live.com> - 2012-09-23 20:42 +0300
      Re: Fastest web framework Stefan Behnel <stefan_ml@behnel.de> - 2012-09-23 19:48 +0200
      RE: Fastest web framework Andriy Kornatskyy <andriy.kornatskyy@live.com> - 2012-09-23 20:57 +0300
      RE: Fastest web framework Andriy Kornatskyy <andriy.kornatskyy@live.com> - 2012-09-23 21:15 +0300
      RE: Fastest web framework Andriy Kornatskyy <andriy.kornatskyy@live.com> - 2012-09-23 21:16 +0300
      RE: Fastest web framework Andriy Kornatskyy <andriy.kornatskyy@live.com> - 2012-09-23 21:17 +0300
      RE: Fastest web framework Andriy Kornatskyy <andriy.kornatskyy@live.com> - 2012-09-23 21:41 +0300

#29800 — Fastest web framework

I have run recently a benchmark of a trivial 'hello world' application for various python web frameworks (bottle, django, flask, pyramid, web.py, wheezy.web) hosted in uWSGI/cpython2.7 and gunicorn/pypy1.9... you might find it interesting:

http://mindref.blogspot.com/2012/09/python-fastest-web-framework.html

Comments or suggestions are welcome.

Thanks.

Andriy Kornatskyy
 		 	   		  

[toc] | [next] | [standalone]

#29813

In article <mailman.1110.1348392023.27098.python-list@python.org>,
 Andriy Kornatskyy <andriy.kornatskyy@live.com> wrote:

> I have run recently a benchmark of a trivial 'hello world' application for 
> various python web frameworks (bottle, django, flask, pyramid, web.py, 
> wheezy.web) hosted in uWSGI/cpython2.7 and gunicorn/pypy1.9... you might find 
> it interesting:
> 
> http://mindref.blogspot.com/2012/09/python-fastest-web-framework.html
> 
> Comments or suggestions are welcome.

That's a nice comparison, thanks for posting it.

One thing that's worth pointing out, however, is that in a real world 
application, as long as you're using something halfway decent, the speed 
of the framework is probably not going to matter at all.  It's much more 
likely that database throughput will be the dominating factor.

[toc] | [prev] | [next] | [standalone]

#29819

Roy Smith, 23.09.2012 16:02:
> Andriy Kornatskyy wrote:
>> I have run recently a benchmark of a trivial 'hello world' application for 
>> various python web frameworks (bottle,�django, flask, pyramid, web.py, 
>> wheezy.web) hosted in uWSGI/cpython2.7 and gunicorn/pypy1.9... you might find 
>> it interesting:
>>
>> http://mindref.blogspot.com/2012/09/python-fastest-web-framework.html
>>
>> Comments or suggestions are welcome.
> 
> That's a nice comparison, thanks for posting it.
> 
> One thing that's worth pointing out, however, is that in a real world 
> application, as long as you're using something halfway decent, the speed 
> of the framework is probably not going to matter at all.  It's much more 
> likely that database throughput will be the dominating factor.

Yes, that makes the comparison (which may or may not be biased towards his
own engine) a bit less interesting. Worth keeping this in mind:

http://www.codeirony.com/?p=9

Stefan

[toc] | [prev] | [next] | [standalone]

#29829

On 23/09/2012 16:50, Stefan Behnel wrote:
> Roy Smith, 23.09.2012 16:02:
>> Andriy Kornatskyy wrote:
>>> I have run recently a benchmark of a trivial 'hello world' application for
>>> various python web frameworks (bottle,�django, flask, pyramid, web.py,
>>> wheezy.web) hosted in uWSGI/cpython2.7 and gunicorn/pypy1.9... you might find
>>> it interesting:
>>>
>>> http://mindref.blogspot.com/2012/09/python-fastest-web-framework.html
>>>
>>> Comments or suggestions are welcome.
>>
>> That's a nice comparison, thanks for posting it.
>>
>> One thing that's worth pointing out, however, is that in a real world
>> application, as long as you're using something halfway decent, the speed
>> of the framework is probably not going to matter at all.  It's much more
>> likely that database throughput will be the dominating factor.
>
> Yes, that makes the comparison (which may or may not be biased towards his
> own engine) a bit less interesting. Worth keeping this in mind:
>
> http://www.codeirony.com/?p=9
>
> Stefan
>
>

I'd like to say thanks for the link but unfortunately for me, but good 
news for you (plural), is that I've bust a gut laughing out loud, so I 
won't :)

Oh alright then thanks for the link.

-- 
Cheers.

Mark Lawrence.

[toc] | [prev] | [next] | [standalone]

#29831

If we take a look at web application we can split it into at least two parts, one that renders things out and the other one that does data extraction, e.g. from database (this is what you are pointing at).

If you made a first call to database you get your list and can easily cache it. The next call IS without impact that database call may cause... but you still keep serving pages out...

Thanks.

Andriy


----------------------------------------
From: roy@panix.com
Subject: Re: Fastest web framework
Date: Sun, 23 Sep 2012 10:02:28 -0400
To: python-list@python.org


In article <mailman.1110.1348392023.27098.python-list@python.org>,
Andriy Kornatskyy <andriy.kornatskyy@live.com> wrote:

> I have run recently a benchmark of a trivial 'hello world' application for
> various python web frameworks (bottle, django, flask, pyramid, web.py,
> wheezy.web) hosted in uWSGI/cpython2.7 and gunicorn/pypy1.9... you might find
> it interesting:
>
> http://mindref.blogspot.com/2012/09/python-fastest-web-framework.html
>
> Comments or suggestions are welcome.

That's a nice comparison, thanks for posting it.

One thing that's worth pointing out, however, is that in a real world
application, as long as you're using something halfway decent, the speed
of the framework is probably not going to matter at all. It's much more
likely that database throughput will be the dominating factor.

--
http://mail.python.org/mailman/listinfo/python-list
 		 	   		  

[toc] | [prev] | [next] | [standalone]

#29833

Andriy Kornatskyy, 23.09.2012 19:42:
> If we take a look at web application we can split it into at least two
> parts, one that renders things out and the other one that does data
> extraction, e.g. from database (this is what you are pointing at).
> 
> If you made a first call to database you get your list and can easily
> cache it. The next call IS without impact that database call may
> cause... but you still keep serving pages out...

Well, if it was really that easy, you wouldn't be using a database in the
first place but static pages, would you?

Stefan

[toc] | [prev] | [next] | [standalone]

#29837

Few facts that doesn't make it less interesting:

(1) the test source code available
(2) the test itself is pretty famous
(3) you can re-run it
(4) or even better supply own that in your believe is 100% relevant

Not every project has problem with database performance. Some use caching... and pretty happy. In my case I have got 2x boost of web application performance just by switching to wheezy.template, that simple.

Thanks.

Andriy


----------------------------------------
> To: python-list@python.org
> From: stefan_ml@behnel.de
> Subject: Re: Fastest web framework
> Date: Sun, 23 Sep 2012 17:50:20 +0200
>
> Roy Smith, 23.09.2012 16:02:
> > Andriy Kornatskyy wrote:
> >> I have run recently a benchmark of a trivial 'hello world' application for
> >> various python web frameworks (bottle,�django, flask, pyramid, web.py,
> >> wheezy.web) hosted in uWSGI/cpython2.7 and gunicorn/pypy1.9... you might find
> >> it interesting:
> >>
> >> http://mindref.blogspot.com/2012/09/python-fastest-web-framework.html
> >>
> >> Comments or suggestions are welcome.
> >
> > That's a nice comparison, thanks for posting it.
> >
> > One thing that's worth pointing out, however, is that in a real world
> > application, as long as you're using something halfway decent, the speed
> > of the framework is probably not going to matter at all. It's much more
> > likely that database throughput will be the dominating factor.
>
> Yes, that makes the comparison (which may or may not be biased towards his
> own engine) a bit less interesting. Worth keeping this in mind:
>
> http://www.codeirony.com/?p=9
>
> Stefan
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
 		 	   		  

[toc] | [prev] | [next] | [standalone]

#29839

Good to know you are in a good humor today. You will be surprised... far not all share your point of view. ;-)

Few links for you to stop laughing that loud:
http://packages.python.org/wheezy.http/userguide.html#content-cache
http://packages.python.org/wheezy.caching/userguide.html#cachedependency

Andriy


----------------------------------------
> To: python-list@python.org
> From: breamoreboy@yahoo.co.uk
> Subject: Re: Fastest web framework
> Date: Sun, 23 Sep 2012 18:20:03 +0100
>
> On 23/09/2012 16:50, Stefan Behnel wrote:
> > Roy Smith, 23.09.2012 16:02:
> >> Andriy Kornatskyy wrote:
> >>> I have run recently a benchmark of a trivial 'hello world' application for
> >>> various python web frameworks (bottle,�django, flask, pyramid, web.py,
> >>> wheezy.web) hosted in uWSGI/cpython2.7 and gunicorn/pypy1.9... you might find
> >>> it interesting:
> >>>
> >>> http://mindref.blogspot.com/2012/09/python-fastest-web-framework.html
> >>>
> >>> Comments or suggestions are welcome.
> >>
> >> That's a nice comparison, thanks for posting it.
> >>
> >> One thing that's worth pointing out, however, is that in a real world
> >> application, as long as you're using something halfway decent, the speed
> >> of the framework is probably not going to matter at all. It's much more
> >> likely that database throughput will be the dominating factor.
> >
> > Yes, that makes the comparison (which may or may not be biased towards his
> > own engine) a bit less interesting. Worth keeping this in mind:
> >
> > http://www.codeirony.com/?p=9
> >
> > Stefan
> >
> >
>
> I'd like to say thanks for the link but unfortunately for me, but good
> news for you (plural), is that I've bust a gut laughing out loud, so I
> won't :)
>
> Oh alright then thanks for the link.
>
> --
> Cheers.
>
> Mark Lawrence.
>
> --
> http://mail.python.org/mailman/listinfo/python-list
 		 	   		  

[toc] | [prev] | [next] | [standalone]

#29840

Good to know you are in a good humor today. You will be surprised... far not all share your point of view. ;-)

Few links for you to stop laughing that loud:
http://packages.python.org/wheezy.http/userguide.html#content-cache
http://packages.python.org/wheezy.caching/userguide.html#cachedependency

Andriy


----------------------------------------
> To: python-list@python.org
> From: breamoreboy@yahoo.co.uk
> Subject: Re: Fastest web framework
> Date: Sun, 23 Sep 2012 18:20:03 +0100
>
> On 23/09/2012 16:50, Stefan Behnel wrote:
> > Roy Smith, 23.09.2012 16:02:
> >> Andriy Kornatskyy wrote:
> >>> I have run recently a benchmark of a trivial 'hello world' application for
> >>> various python web frameworks (bottle,�django, flask, pyramid, web.py,
> >>> wheezy.web) hosted in uWSGI/cpython2.7 and gunicorn/pypy1.9... you might find
> >>> it interesting:
> >>>
> >>> http://mindref.blogspot.com/2012/09/python-fastest-web-framework.html
> >>>
> >>> Comments or suggestions are welcome.
> >>
> >> That's a nice comparison, thanks for posting it.
> >>
> >> One thing that's worth pointing out, however, is that in a real world
> >> application, as long as you're using something halfway decent, the speed
> >> of the framework is probably not going to matter at all. It's much more
> >> likely that database throughput will be the dominating factor.
> >
> > Yes, that makes the comparison (which may or may not be biased towards his
> > own engine) a bit less interesting. Worth keeping this in mind:
> >
> > http://www.codeirony.com/?p=9
> >
> > Stefan
> >
> >
>
> I'd like to say thanks for the link but unfortunately for me, but good
> news for you (plural), is that I've bust a gut laughing out loud, so I
> won't :)
>
> Oh alright then thanks for the link.
>
> --
> Cheers.
>
> Mark Lawrence.
>
> --
> http://mail.python.org/mailman/listinfo/python-list
 		 	   		  

[toc] | [prev] | [next] | [standalone]

#29841

Good to know you are in a good humor today. You will be surprised... far not all share your point of view. ;-)

Few links for you to stop laughing that loud:
http://packages.python.org/wheezy.http/userguide.html#content-cache
http://packages.python.org/wheezy.caching/userguide.html#cachedependency

Andriy


----------------------------------------
> To: python-list@python.org
> From: breamoreboy@yahoo.co.uk
> Subject: Re: Fastest web framework
> Date: Sun, 23 Sep 2012 18:20:03 +0100
>
> On 23/09/2012 16:50, Stefan Behnel wrote:
> > Roy Smith, 23.09.2012 16:02:
> >> Andriy Kornatskyy wrote:
> >>> I have run recently a benchmark of a trivial 'hello world' application for
> >>> various python web frameworks (bottle,�django, flask, pyramid, web.py,
> >>> wheezy.web) hosted in uWSGI/cpython2.7 and gunicorn/pypy1.9... you might find
> >>> it interesting:
> >>>
> >>> http://mindref.blogspot.com/2012/09/python-fastest-web-framework.html
> >>>
> >>> Comments or suggestions are welcome.
> >>
> >> That's a nice comparison, thanks for posting it.
> >>
> >> One thing that's worth pointing out, however, is that in a real world
> >> application, as long as you're using something halfway decent, the speed
> >> of the framework is probably not going to matter at all. It's much more
> >> likely that database throughput will be the dominating factor.
> >
> > Yes, that makes the comparison (which may or may not be biased towards his
> > own engine) a bit less interesting. Worth keeping this in mind:
> >
> > http://www.codeirony.com/?p=9
> >
> > Stefan
> >
> >
>
> I'd like to say thanks for the link but unfortunately for me, but good
> news for you (plural), is that I've bust a gut laughing out loud, so I
> won't :)
>
> Oh alright then thanks for the link.
>
> --
> Cheers.
>
> Mark Lawrence.
>
> --
> http://mail.python.org/mailman/listinfo/python-list
 		 	   		  

[toc] | [prev] | [next] | [standalone]

#29845

The problem is that easy if you have a complete control over what you are caching.

Complete control over cache may look a challenging task however with use of cache dependency you can lower it significantly. Take a look here:
http://packages.python.org/wheezy.caching/userguide.html#cachedependency

If you have a willing to go even further consider take a look at content caching:
http://packages.python.org/wheezy.http/userguide.html#content-cache

Serving static page out of your data is not that impossible... there are still exceptions, of cause.

Thanks.

Andriy


----------------------------------------
> To: python-list@python.org
> From: stefan_ml@behnel.de
> Subject: Re: Fastest web framework
> Date: Sun, Sep 2 :::::: +0<<<
>
> Andriy Kornatskyy, ......2 ::::::
> > If we take a look at web application we can split it into at least two
> > parts, one that renders things out and the other one that does data
> > extraction, e.g. from database (this is what you are pointing at).
> >
> > If you made a first call to database you get your list and can easily
> > cache it. The next call IS without impact that database call may
> > cause... but you still keep serving pages out...
>
> Well, if it was really that easy, you wouldn't be using a database in the
> first place but static pages, would you?
>
> Stefan
>
> --
> http://mail.python.org/mailman/listinfo/python-list
 		 	   		  

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.python

csiph-web