Re: 64-bit CDB

Matt Chaput <[email protected]> Fri, 19 Mar 2010 13:27:50 -0400

Newsgroups	gmane.comp.djb.cdb
Message-ID	<[email protected]>

On 3/19/2010 11:51 AM, Matt Chaput wrote:
> Let me try to be clearer: does anyone understand the hashtable lookup
> part of the CDB algorithm (i.e. start at (hash >> 8) % tablesize, half
> the table kept empty, etc.) sufficiently to explain if it could/how it
> would change if I use 12-byte records (4-byte hash + 8-byte pointer)?

Sorry, y'know what, I think I grok it now. Thanks anyway :)

Matt