Re: [PERFORM] "Hash index" vs. "b-tree index" (PostgreSQL

Added to TODO:

* Consider sorting hash buckets so entries can be found using a binary
search, rather than a linear scan
* In hash indexes, consider storing the hash value with or instead
of the key itself

---------------------------------------------------------------------------

Neil Conway wrote:
> Bruce Momjian wrote:
> > Is there a TODO anywhere in this discussion? If so, please let me know.
>
> There are a couple:
>
> - consider changing hash indexes to keep the entries in a hash bucket
> sorted, to allow a binary search rather than a linear scan
>
> - consider changing hash indexes to store each key's hash value in
> addition to or instead of the key value.
>
> You should probably include a pointer to this discussion as well.
>
> (I'd like to take a look at implementing these if I get a chance.)
>
> -Neil
>

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073