I am sorry for not elaborating on that. What I meant by de-correlation was
optimizing a query to get rid of sub-queirs by using joins.
eg. In the TPC-H schema, a query to find out the names of suppliers who
supply parts having size < 100
*Query with nested subqueries:*
SELECT
S_NAME
FROM
SUPPLIER
WHERE
S_SUPPKEY
IN (
SELECT
PS_SUPPKEY
FROM
PARTSUPP
WHERE
PS_PARTKEY
IN (
SELECT
P_PARTKEY
FROM
PART
WHERE
P_SIZE < 100
)
*Query with joins without subqueries:*
SELECT
S_NAME
FROM
SUPPLIER
INNER JOIN
PARTSUPP
ON
S_SUPPKEY = PS_SUPPKEY
INNER JOIN
PART
ON
P_PARTKEY = PS_PARTKEY
WHERE
P_SIZE < 100
Thanks,
Mahendra
On Thu, Jul 23, 2009 at 9:02 PM, Itagaki Takahiro <
[email protected]> wrote:
>
> mahendra chavan <[email protected]> wrote:
>
> > I am a master's student in computer science at IIT Bombay. As part of my
> > project, I need to get a decorrelated version of a SQL query. Please
> could
> > anyone let me know if we have query decorrelation feature implemented in
> > postgres ?
>
> What do you mean by "query decorrelation"? Is it an addtional method for
> query optimization? At least there is no word 'decorrelation' in
> the postgres documentation.
>
> Regards,
> ---
> ITAGAKI Takahiro
> NTT Open Source Software Center
>
>
>
