Digest::Pearson is an implementation of Peter K. Pearson's hash algorithm
presented in "Fast Hashing of Variable Length Text Strings" - ACM 1990. This
hashing technique yields good distribution of hashed results for variable
length input strings on the range 0-255, and thus, it is well suited for
data load balancing.
The implementation is in C, so it is fast. If you prefer a pure Perl version
and can tolerate slower speed, you might want to consider
Digest::Pearson::PurePerl instead.

WWW: http://search.cpan.org/dist/Digest-MD5-Pearson/