On Fri, 8 Apr 2011 16:07:11 -0500 Kenneth Marshall <k...@rice.edu> wrote:
[...] > The question > is are the collisions with the hash function used in DSPAM impacting its > effectiveness to a degree in which it would make sense to change the > function? > Short anwser: no Longer answer: Effectiveness is not only about how collision resistent a hashing function is. Speed, memory usage, etc... all this counts too. Beside that the current implentation in DSPAM is not among the fastest CRC64 algorithms. Intel has a great paper on how to speed up CRC computing on their arch -> http://www.intel.com/technology/comms/perfnet/download/CRC_generators.pdf <- using a pure software approach. One could use the technique outlined in the PDF from above and still keep the same old polynomial DSPAM uses and prouce faster a CRC64-ISO hash. I personally would add a configuration option either in dspam.conf or in the preference extension allowing the DSPAM operator to choose between CRC64-ISO (current DSPAM hashing algorithm) and something like CRC64-ECMA-182 or any other (future) hashing algorithm. > Cheers, > Ken > -- Kind Regards from Switzerland, Stevan Bajić ------------------------------------------------------------------------------ Xperia(TM) PLAY It's a major breakthrough. An authentic gaming smartphone on the nation's most reliable network. And it wants your games. http://p.sf.net/sfu/verizon-sfdev _______________________________________________ Dspam-user mailing list Dspam-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dspam-user
