On Mon, Jul 08, 2002 at 05:24:28AM -0700, Thirumoorthy Bhuvneswari wrote:
> I am using Postgresql-7.1.3 with RedHat Linux-7.1 and
> JDBC. I am having an indexed table with nearly 3,000
> records for a month.
How many months are you storing? i.e. how large is the
entire table?
> If I run the query from the hard-disk it takes about
> 5 seconds to display all the 3,000 records. If I
> connect 3 nodes to it and run the query from a node,
> it takes about 1.5 minutes.
Not sure what you mean by "running it from the hard disk",
versus "connecting nodes". If you mean that the query takes
longer when executed by clients on a LAN, sounds like a
network problem.
If not, then you need to provide a LOT more information. As is,
we can't help you. Have you VACUUM ANALYZE'd the tables in
question? How much data (# of rows) is involved? What queries
are being executed? What does EXPLAIN show for these queries?
What is the definition of the tables involved, as well as
any applicable indexes? Under what exact circumstances does
the performance problem occur?
Cheers,
Neil
--
Neil Conway <neilconway@rogers.com>
PGP Key ID: DB3C29FC