[OSM-dev] osm2pgsql-64 and benchmarks

Martijn van Oosterhout kleptog at gmail.com
Wed Jun 22 07:31:18 BST 2011


On 21 June 2011 13:44, Frederik Ramm <frederik at remote.org> wrote:

>   as you know I've recently made the osm2pgsql-64 branch which does not use
> intarray. The osm2pgsql-64 branch is capable of handling 64-bit IDs but can
> also be defined to use 32-bit IDs like the classic osm2pgsql.

Cool.

> The results (all times in minutes):
>
>                                  Postgres 8.4    Postgres 9.0
> --------------------------------------------------------------------
> old osm2pgsql (intarray)              401              308
> osm2pgsql-64 with 32bit IDs           407              297
> osm2pgsql-64 with 64bit IDs           415              302

Those are significant differences there. Given there queries we do I
wouldn't have expected pass-by-val to make much difference, there's no
calculations happening so you don't save any alloc overhead.

The reason I went for arrays of integers is diskspace (4 bytes per
node rather than 40 for a whole row). Did you compare that?

Postgres gets faster every release :) I wouldn't be able suggest which
bit is helping here...

Mvg,
-- 
Martijn van Oosterhout <kleptog at gmail.com> http://svana.org/kleptog/



More information about the dev mailing list