Re: SQL challenge;-)

On Thu, 3 May 2001 00:20:01 +0200, "Frank an der Heiden" <fh_at_energotec.de> wrote:

>hi,
>
>here a little task for all sql-freaks ;-))
>
>i have to tables.
>these tables have the same structure.
>i want to check out which row of <column> of the first table is not
>existing in <column> of the second table.
>
>this sql-statement works:
>
>SELECT <column> FROM <table1> WHERE <column> NOT IN (SELECT <column> FROM
><table2>);
>
>but it is very very slow, because the second select is done very often if
>you have many rows in table1.
>any simple idea for improving performance?
>a tried to use a join, but i wasn't able to bring it to work, maybe it's too
>late ;-((
>
>good night
>frank
>
>

SELECT <column> FROM <table1> WHERE <column> NOT IN (/*+HASH_AJ(table2) */ SELECT <column> FROM <table2>);

Just my 2 eurocents

Good night

Sybrand Bakker, Oracle DBA Received on Wed May 02 2001 - 18:24:13 CDT