On Mon, 2008-08-18 at 09:46 +0800, Xiao Meng wrote:
> There's minor change against the previous
> one( http://archives.postgresql.org/pgsql-hackers/2008-07/msg01183.php ).
> * merge branch master(Aug 16) into the patch
> * clean code and make some comment
> Performance result is here
> http://wiki.postgresql.org/wiki/Gsoc08-hashindex
>
> It seems hash index is a little better on index creation and
> selection.
> But maybe it's in the range of noise, I'm not sure.
> I'd like to try it with a bigger dataset (e.g. table with 10GB) but
> there is not enough space in my computer.
> Anyone interest can make a test on a bigger data set.
You don't give the text of the query used to do these performance tests,
so I can't validate your test results.
Right now it seems strange that the index is larger than a btree, yet
the performance tests show that 3 times as much I/O was used accessing
the btree.
--
Simon Riggs www.2ndQuadrant.com
PostgreSQL Training, Services and Support