[Haskell-cafe] Re: Haskell Speed

[Paul Moore]

On 23 Dec 2005 22:29:02 +0100, Peter Simons <simons at cryp.to> wrote:
> Daniel Carrera writes:
>
>  > when I have a simple algorithm and performance is an
>  > issue [...] I'd use C.
>
> You don't have to. You can write very fast programs in
> Haskell.
>
> I never really finished the article I wanted to write about
> this subject, but the fragment I have might be interesting
> or even useful nonetheless:
>
>   http://cryp.to/blockio/fast-io.html
>   http://cryp.to/blockio/fast-io.lhs

One of the interesting points that this illustrates (to me) is that
the "obvious" approach in Haskell can be seriously non-optimal in
terms of performance. Add to this the fact that tuning functional
programs is a non-trivial art, and it becomes quite easy to see why
Haskell might come across as slow.

Your wcLazy implementation is (as you say) the "obvious" approach.
Your two other approaches aren't particularly obvious, althought they
are far more efficient. It would be interesting to see standalone code
for wcIOB (where you're allowed to assume that any helpers you need,
like your block IO library, are available from the standard library).
This would help in comparing the "obviousness" of the two approaches.

Paul