[IronPython] pybench results for CPython and IronPython

Jim Hugunin Jim.Hugunin at microsoft.com
Sun Apr 22 20:50:41 CEST 2007

Hi Seo,

I saw nothing wrong with your original post.  I had just noticed that others on the list were taking these numbers as indicative of expected IronPython performance anywhere and I wanted to be sure that they could also see the numbers on .NET for the rest of the story.

Trying to understand these numbers and use the relative performance to help improve all of the different implementations out there can only be a good thing.  Often these kinds of discrepancies can help identify low-hanging fruit for performance optimizations.  I know that we're going to be looking at our dictionaries and our slicing operation in much more detail during the next performance push.  I also remember that IronPython-0.6's poor performance on try/raise/except was noted by the Mono team - at the time, Mono was 10x slower than the already slow .NET here.  I believe that this was the original motivation for the performance improvements to that Mono path that now make it even faster than .NETs.

In the spirit of more data, I've attached my raw results files for you in case they help with your comparisons.  My results are on a ThinkPad X60 laptop with an intel T2400 1.83 GHz core duo processor and 1.5GB of RAM running Windows Vista.

Thanks - Jim

-----Original Message-----
From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of Sanghyeon Seo
Sent: Sunday, April 22, 2007 7:35 AM
To: Discussion of IronPython
Subject: Re: [IronPython] pybench results for CPython and IronPython

2007/4/22, Jim Hugunin <Jim.Hugunin at microsoft.com>:
> I'd like to point out one fact about these benchmarks that seems to have been missed.  Seo's numbers are for running IronPython 1.1 on Mono - not on the Microsoft .NET implementation.  The Mono team has done excellent work with that project, but today the performance of the .NET implementation is still significantly better and is a much better platform for benchmarking IronPython's performance.

I never tried to claim otherwise. Perhaps I should have been more explicit.

> I reran pybench on my laptop tonight on the RTM version of .NET that ships with Windows Vista - and has been available for XP for quite some time now.  I compared with CPython-2.5.1 (the latest version).
> (snip)

Can you provide the raw pybench result files? I can host it on my site
side-by-side with Mono results.

Seo Sanghyeon
users mailing list
users at lists.ironpython.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: py25.pybench
Type: application/octet-stream
Size: 49233 bytes
Desc: py25.pybench
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20070422/55f5d9c3/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ipy11.pybench
Type: application/octet-stream
Size: 44667 bytes
Desc: ipy11.pybench
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20070422/55f5d9c3/attachment-0001.obj>

More information about the Ironpython-users mailing list