On Thu, Jun 27, 2002 at 01:46:28PM -0700, Don Cohen wrote: > > I had mentioned that I was hoping to remember/find the appropriate > formula for the probability of a hash bucket reaching its limit. > I got out my prob-stat book from I-won't-mention-how-many years ago.
[thanks for your detailad analysis, I will try to look at all the details ASAP]. > My guess is that the amount of time required to do 20 iterations > comparing connection data with elements of a list in a bucket is > very small compared to the amount of time otherwise spent on a > packet. On the other hand, a hash bucket is also pretty cheap. > The data for a connection takes much more space than 2 hash buckets, > or even 10. And with 10L buckets the chance of overflow at M=10 > is only 1E-19, for M=20 it's somewhere around 1E-40. ok the question is now to test this in the real world, which can be fairly easily done, as the hashsize is configurable. > So, there you have it. -- Live long and prosper - Harald Welte / [EMAIL PROTECTED] http://www.gnumonks.org/ ============================================================================ GCS/E/IT d- s-: a-- C+++ UL++++$ P+++ L++++$ E--- W- N++ o? K- w--- O- M- V-- PS+ PE-- Y+ PGP++ t++ 5-- !X !R tv-- b+++ DI? !D G+ e* h+ r% y+(*)
