Final issues to fix for the shootout entries (Was: no subject)

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Final issues to fix for the shootout entries (Was: no subject)

Donald Bruce Stewart
Yeah, have a look on the shootout:

    http://shootout.alioth.debian.org/gp4/benchmark.php?test=recursive&lang=all

Has an ackerman.

More detailed comparisions, across the 19 benchmarks on the shootout,
with notes on which of the 19 GHC Haskell is significantly slower (>4x) at:

    Haskell verus Python    
        http://shootout.alioth.debian.org/gp4/benchmark.php?test=all&lang=ghc&lang2=python
        Loses on:
            regex-dna

    Haskell verus Ruby
        http://shootout.alioth.debian.org/gp4/benchmark.php?test=all&lang=ghc&lang2=ruby
        Loses on:
            regex-dna

    Haskell verus Perl
        http://shootout.alioth.debian.org/gp4/benchmark.php?test=all&lang=ghc&lang2=perl
        Loses on:
            regex-dna

    Haskell verus Lisp
        http://shootout.alioth.debian.org/gp4/benchmark.php?test=all&lang=ghc&lang2=sbcl
        Loses on:
            mandelbrot (Double math)
            spectral-norm (Double math)

    Haskell verus Clean
        http://shootout.alioth.debian.org/gp4/benchmark.php?test=all&lang=ghc&lang2=clean
        Loses on:
            k-nucleotide ( HashTable benchmark)
            mandelbrot
            spectral-norm

    Haskell verus SML
        http://shootout.alioth.debian.org/gp4/benchmark.php?test=all&lang=ghc&lang2=mlton
        Loses on:
            k-nucleotide
            mandelbrot
            spectral-norm

    Haskell versus OCaml
        http://shootout.alioth.debian.org/gp4/benchmark.php?test=all&lang=ghc&lang2=ocaml
        Loses on:
            k-nucleotide
            mandelbrot
            regex-dna
            spectral-norm

    Haskell versus C
        Loses on:
            k-nucleotide
            mandelbrot
            regex-dna
            spectral-norm
           
Fixing these last few programs:

    k-nucleotide:       Needs a better hashtable library implementation. For
                        example a Data.HashByteString based on a Trie of ByteString)

    regex-dna:          Should use the regex-tre library instead of
                        regex-posix (algorithm issue only)

    mandelbort/spectral-norm:   Double math is slow. I don't know why.

And note that strings aren't a bottleneck anymore!

    http://shootout.alioth.debian.org/gp4/benchmark.php?test=sumcol&lang=all

-- Don

seth:

> GHC has profiling support.
>
> (By the way, many mail servers these days discard mail with no subject.)
>
> I've seen a number of papers comparing the speed of Haskell code to code of other functional languages; there is a periodic "shoot out" with ocaml.
>
> Some probably have comparisons with imperative languages, and, even if they do not, the methodology should help you.
>
> Seth Kurtzberg
>
> On Mon, 5 Feb 2007 11:28:03 -0800 (PST)
> Tays Soares <[hidden email]> wrote:
>
> > Hello everyone,
> >
> > I did at my master thesis a compiler that generates Haskell code. Now I need to measure the execution time of my generated code and I've been searched and I don't know if I'm looking with the wrong keywords but I could not find anything. I just need to measure the time of simple functions, like Ackermann and Fibonacci. Does anyone know how to measure the execution time of a Haskell program or function?
> >
> > Thank you,
> > Tays
> >
> >
> >
> > __________________________________________________
> > Fale com seus amigos  de gra?a com o novo Yahoo! Messenger
> > http://br.messenger.yahoo.com/
> _______________________________________________
> Glasgow-haskell-users mailing list
> [hidden email]
> http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
_______________________________________________
Haskell-Cafe mailing list
[hidden email]
http://www.haskell.org/mailman/listinfo/haskell-cafe
Reply | Threaded
Open this post in threaded view
|

Re: Final issues to fix for the shootout entries (Was: no subject)

Tomasz Zielonka
On Tue, Feb 06, 2007 at 12:12:09PM +1100, Donald Bruce Stewart wrote:
> And note that strings aren't a bottleneck anymore!
>
>     http://shootout.alioth.debian.org/gp4/benchmark.php?test=sumcol&lang=all

I think "Haskell GHC #2" doesn't play by the rules:

    "
    diff program output for this 6KB input file with this output file to
    check your program is correct before contributing.

    Each program should be implemented the same way as this Icon program.

    The sum-file benchmark measures line-oriented I/O and string conversion.

    Each program should:

        * read integers from stdin, one line at a time
        * print the sum of those integers

    Programs should use built-in line-oriented I/O functions rather than
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    custom-code. No line will exceed 128 characters, including newline.
    ^^^^^^^^^^^^
    Reading one line at a time, the programs should run in constant space.
    "

Best regards
Tomasz
_______________________________________________
Haskell-Cafe mailing list
[hidden email]
http://www.haskell.org/mailman/listinfo/haskell-cafe