Posted by Erland Sommarskog on 11/29/99 11:52

Thomas R. Hummel (tom_hummel@hotmail.com) writes:
> Thanks for the interesting solution Erland. In the real life case of
> course, I will have many more columns than just my_id and line_number.
> I realize that everything is dependent on the specific data, etc., but
> in general, would you expect this to perform better than:

I will have to admit not having done any benchmark on the two solutions.
But my gut feeling is that the solution with MIN and GROUP BY generally
is better. The other seems to give more chances for the optimizer to go
astray.

--
Erland Sommarskog, SQL Server MVP, esquel@sommarskog.se

Books Online for SQL Server 2005 at
http://www.microsoft.com/technet/prodtechnol/sql/2005/downloads/books.mspx
Books Online for SQL Server 2000 at
http://www.microsoft.com/sql/prodinfo/previousversions/books.mspx

• Next in forum: Re: Make and save a blob on the fly in stored proc
• Prev in forum: Re: Strange results from "not in" query
• Thread view: Re: Need help with denormalizing query

[Reply to this message]