NOTE: use Perl; is on undef hiatus. You can read content, but you can't post it. More info will be forthcoming forthcomingly.
All the Perl that's Practical to Extract and Report
use Perl Log In
djberg96 (2603)
djberg96
(email not shown publicly)
(email not shown publicly)
Wednesday May 28, 2003
02:40 PM
Benchmarking
I'm into benchmarks. In the world of TMTOWTDI, I like to know which way is the *fastest*. The I've run various benchmarks for both Perl and Ruby, but it was always just comparing a couple methods here, a couple methods there. I never put together any sort of formal benchmarking suite.
Now, I'm aware of The Great Computer Language Shootout. It's nice for seeing how language X stacks up against language Y. But it's NOT good for comparing language X against itself.
So, I decided to put together a benchmarksuite to see how various operations compare from 1.6.x. to 1.8.0 and from machine to machine. I started with Ruby because, in my mind, it's more of an issue for Ruby than it is for Perl. In addition, things seem to be getting *slower* from 1.6.x to 1.8.0 and I wanted some hard numbers to back up my concerns.
I've posted the results here if you're interested. See the "Thoughts and Musings" at the bottom as well.
Anyone be interested in something like this for Perl? I didn't really see anything on CPAN.
Benchmarking
More |
Login
| Reply
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
I pour myself a cup of strong coffee. :-)
-- Jarkko Hietaniemi
Stories, comments, journals, and other submissions on use Perl; are Copyright 1998-2006, their respective owners.
wantead - dead or alive - benchmarks (Score:3, Informative)
<blink>yes please</blink>.We (p5p) are looking for a decent benchmark to replace perlbench. (Most recently Alan asking about a decent benchmark [mpe.mpg.de]). I've suggested spamassassin [mpe.mpg.de] which would work as a general single number "how fast is perl?", but what we'd really like is a meaningful set of benchmarks where we can see things like "oh, this patch makes OO method calls 181% faster" [mpe.mpg.de] (That was Arthur. Mine are usually "grrr, this patch makes no difference to any perlbench test. Is it me or is it perlbench?")
Reply to This