> -----Original Message-----
> From: pgsql-hackers-owner@hub.org
> [mailto:pgsql-hackers-owner@hub.org]On Behalf Of Mikheev, Vadim
>
> > We might have part of the story in the recently noticed fact that
> > each insert/update query begins by doing a seqscan of pg_index.
> >
> > I have done profiles of INSERT in the past and not found any really
> > spectacular bottlenecks (but I was looking at a test table with no
> > indexes, so I failed to see the pg_index problem :-(). Last time
> > I did it, I had these top profile entries for inserting 100,000 rows
> > of 30 columns apiece:
>
> Well, I've dropped index but INSERTs still take 70 sec and
> COPY just 1sec -:(((
>
Did you run vacuum after dropping indexes ?
Because DROP INDEX doesn't update relhasindex of pg_class,
planner/executer may still look up pg_index.
Regards.
Hiroshi Inoue
Inoue@tpf.co.jp