Tom Lane wrote:
Tatsuo Ishii <[EMAIL PROTECTED]> writes:
BTW, in 8.0 table spaces allow a table and an index to be placed on
different disk devices, which may improve the performance. Is it
possible that planner accounts for this in the future.
I'm not sure how we could model it in any useful way
Tatsuo Ishii <[EMAIL PROTECTED]> writes:
> BTW, in 8.0 table spaces allow a table and an index to be placed on
> different disk devices, which may improve the performance. Is it
> possible that planner accounts for this in the future.
I'm not sure how we could model it in any useful way ...
Sorry for the delay. I was out of town last night.
> What can you tell us about the physical ordering of the tables involved?
> The only way I can think of to explain these results is that "orders"
> is more or less physically in order by its primary key, and "lineitem"
> is also more or less orde
Tatsuo Ishii <[EMAIL PROTECTED]> writes:
> I've been playing with OSDL DBT-3 for a while and found very strange
> phenomemon.
What can you tell us about the physical ordering of the tables involved?
The only way I can think of to explain these results is that "orders"
is more or less physically in
On L, 2004-09-04 at 17:35, Tatsuo Ishii wrote:
> Hi,
>
> I've been playing with OSDL DBT-3 for a while and found very strange
> phenomemon. The query in question is Q4:
>
> select o_orderpriority, count(*) as order_count from orders where
> o_orderdate >= date '1997-10-01' and o_orderdate < (date
Hi,
I've been playing with OSDL DBT-3 for a while and found very strange
phenomemon. The query in question is Q4:
select o_orderpriority, count(*) as order_count from orders where
o_orderdate >= date '1997-10-01' and o_orderdate < (date
'1997-10-01' + interval '3 month')::date and o_orderkey in (
