Re: Speedup the dml ??
Date: Fri, 14 Apr 2000 18:44:20 +0800
Message-ID: <38F6F684.4899_at_yahoo.com>
feib wrote:
> 
> Hello, there
>    I have two tables a,b ( a with 600000 records, b 60000 records)
>    Both a and b have the same columns c1 c2 c3 c4 c5.
>    I want to update a's c5 by b's c5 for the same key value c1-c4.
>    So,I create two unique indexies on a(c1,c2,c3,c4) and b(c1,c2,c3,c4).
> 
>     update a
>          set c5 = ( select c5
>                          from b
>                        where a.c1=b.c1
>                            and a.c2=b.c2
>                            and a.c3=b.c3
>                            and a.c4=b.c4 );
> 
>     Are there any better statement ( like EXISTS,or any hint to speedup) ??
>     Any ideas ??
> 
> Rgds,
> Agi Chen
> 
> 
> 
Try updating the join directly (this will work if you have the appropriate primary keys)
HTH
-- =========================================== Connor McDonald http://www.oracledba.co.uk We are born naked, wet and hungry...then things get worseReceived on Fri Apr 14 2000 - 12:44:20 CEST
