Python and the need for speed

Brecht Machiels brecht__gmane at mos6581.org
Wed Apr 12 11:04:23 EDT 2017


On 2017-04-12 14:46:45 +0000, Michael Torrie said:

> On 04/12/2017 03:33 AM, Brecht Machiels wrote:
>> However, rinohtype is located in a very different niche and it would
>> greatly benefit a lot from a speed boost. Rendering the Sphinx
>> documentation (311 pages) takes almost 10 minutes on my i7, which is
>> simply too long given the available processing power. And yes, I have
>> spent a lot time profiling and optimizing the code. You're always
>> welcome to demonstrate my incompetence by means of a pull request, of
>> course ;-)
> 
> You talked about PyPy before.  Did you try rinohtype on pypy3?

I did. See 
https://bitbucket.org/pypy/pypy/issues/2365/rinohtype-much-slower-on-pypy3 


A long time ago, I also backported rinohtype to Python2 to test with 
PyPy2. See 
https://mail.python.org/pipermail/pypy-dev/2015-August/013767.html

I haven't been able to try the latest PyPy3 release because there is no 
binary for macOS, and building it using pyenv fails 
(https://github.com/pyenv/pyenv/issues/890).

> I tried to download pypy3 just now but the x86_64 binary package isn't
> compatible with my old distro (Centos 7) and I don't quite have time to
> build it from source.  But if I can get that running I would like to try
> your rinoh demos with it and see how it performs.  I've seen some pretty
> good speedups on slow, CPU-intensive python code before with pypy.

It would be great if you could run the benchmark I mention in my first 
link and share the results. Highly appreciated!

Best regards,
Brecht



More information about the Python-list mailing list