"Aryeh M. Friedman" <[EMAIL PROTECTED]> writes: > Dag-Erling Smørgrav <[EMAIL PROTECTED]> writes: > > "Aryeh M. Friedman" <[EMAIL PROTECTED]> writes: > > > All hashs have issues with pooling.... see > > > http://www.burtleburtle.net/bob/hash/index.html... btw it is a > > > old wives tale that the number of buckets should be prime (mostly > > > based on the very weak implementation Knuth offered) > > Not an "old wives' tale", but rather an easy way to implement a > > hash algorithm that is good enough for most simple uses: metric > > modulo table size, where metric is a number derived from the item > > in such a manner as to give a good spread. > Sorry for taking a while to reply.... but the above only applies if > your using a very primitive hash like Knuth's multiplication one....
You are overlooking a very common case: the use of a hash table to implement an in-memory dictionary (aka associative array) where the key is an integer with poor variability in the high-order bits. K % N where K is the key and N is the size of the table requires very little code, is reasonably efficient, and, provided that N is prime, gives a reasonably good spread (excpet in pathological cases where the values of K are clustered around multiples of N). > every modern hash I know of should have 2^k buckets actually for some > k<2^32 [in almost all cases <2^16 except for algorithms like the one I > mentioned I am working on which sets k=n where n=the bit count of the > key]. I certainly hope not. 2^(2^32) is several of billion orders of magnitude more than the number of elementary particles in the known universe (currently estimated at 10^80). Even 2^(2^16) is too big by about sixty thousand orders of magnitude. DES -- Dag-Erling Smørgrav - [EMAIL PROTECTED] _______________________________________________ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to "[EMAIL PROTECTED]"
