Dear Michael,
Is there a concrete performance problem you came across?
In any software project you will have a number of suboptimal decisions
and algorithms around (as I trust you know yourself). Reality dictates
that you fix those first that matter or where an actual problem shows
up.
But thanks for the info.
Regards
Christian
"Michael Neumann" <uu9r@rz.uni-karlsruhe.de> wrote in message
news:<20021205193623.GB29100@rz.uni-karlsruhe.de>...
> (please CC to uu9r@rz.uni-karlsruhe.de, as I'm not subscribed to the
> list).
>
> Hi,
>
> The hash algorithm used in
> mozart-1.2.4.20020829/platform/emulator/hashtbl.cc
> (it is known as the P.J.Weinberg algorithm) is not the best available.
> Those used by e.g. StandardML or OCaml are much better.
> I'd suggest (if there are no other problems I am not aware of) to
choose the
> SDBM hash (used by Sleepycat's Berkley Database), which improves both
performance
> and has less collisions.
>
> One problem the SDBM algorithm (as well as OCamls or SMLs) would
> solve, is that of resizing the hashtable, because it works with powers
> of 2 for modulo and do not require prime numbers. This results in a
> much better performance on some architectures and is simpler to
> implement.
>
> For more information on the algorithms and benchmarks see:
> http://www.fantasy-coders.de/projects/gh/html/x435.html
>
> Regards,
>
> Michael Neumann
> -
> Please send submissions to hackers@mozart-oz.org
> and administriva mail to hackers-request@mozart-oz.org.
> The Mozart Oz web site is at http://www.mozart-oz.org/.
-
Please send submissions to hackers@mozart-oz.org
and administriva mail to hackers-request@mozart-oz.org.
The Mozart Oz web site is at http://www.mozart-oz.org/.