[mercury-users] Re: (the lack of) unique modes

[Ralph Becket]

I recently conducted a performance experiment in Mercury and found (in
agreement with RA O'Keefe's advice in ``The Art of Prolog'', I think)
that simply improving Mercury's memory usage had a dramatic impact on
performance.

Examples include hashing two small numbers into one, rather than using
pairs, not indexing maps etc. using boxed types, moving from maps to
hash tables (really v. easy to do with Mercury's arrays) and so forth.
In this fashion I managed to net a 10-20 fold speed increase with minor
changes to my code.

Cheers,

Ralph
--------------------------------------------------------------------------
mercury-users mailing list
post:  mercury-users at cs.mu.oz.au
administrative address: owner-mercury-users at cs.mu.oz.au
unsubscribe: Address: mercury-users-request at cs.mu.oz.au Message: unsubscribe
subscribe:   Address: mercury-users-request at cs.mu.oz.au Message: subscribe
--------------------------------------------------------------------------