I am answering just for the sake of answering your questions.
What hubert depesz lubaczewski suggested had fixed the problem i had.
I have other queries that need event_id to be the clustered index
Veh_id is spread all over the table. (for veh_id 3 there are no records)
Due to the spread of reco
Thanks, it did help. Now queries run in zero time.
I had thought of doing thatbut since the same configuration was working
ok on MSSQL I thought it should also here.
Now with that index the server query times are a lot faster than MSSQL
without it.
Since it is working I will leave it like that
On 27 Jan 2011, at 15:04, Michael Kemanetzis wrote:
> Hello, I'm experiencing a strange behavior running a simple select query on a
> table that has about 12 million rows. Specifically, changing the "LIMIT"
> value seems to change the execution plan but the result in one of the cases
> is unjus
On Thu, Jan 27, 2011 at 04:04:02PM +0200, Michael Kemanetzis wrote:
> Hello, I'm experiencing a strange behavior running a simple select query on
> a table that has about 12 million rows. Specifically, changing the "LIMIT"
> value seems to change the execution plan but the result in one of the case
Hello, I'm experiencing a strange behavior running a simple select query on
a table that has about 12 million rows. Specifically, changing the "LIMIT"
value seems to change the execution plan but the result in one of the cases
is unjustifiably slow, as if it ignores all indexes.
The table structur
