December 13, 2007 at 2:39 pm
Now in case of FASTFIRSTROW..I am trying to achieve a fast response time
It's far better to write good fast SQL with the proper indexes instead of trying to trick Mother Nature into doing optimization for you.
--Jeff Moden
Change is inevitable... Change for the better is not.
December 14, 2007 at 5:14 am
sql 2005 doesn't appear to be quite so good at deciding if it should recompile a plan, although I've no direct tests to prove this only seeing what happened to Navision.
Microsoft issued service packs to remove the fast statements from the Navision application becuase it did so much harm - we had over 500 plan guides forcing recompiles whilst we waited for the sp patch. Of course the constant recompiles caused other problems ( oh joy! ) if nothing else this should indicate the risk of using FAST - not every change or additional feature to sql server is always for the best, and quite often there's a "dark side" to such chnages, such as for FAST , varchar(max) and such.
[font="Comic Sans MS"]The GrumpyOldDBA[/font]
www.grumpyolddba.co.uk
http://sqlblogcasts.com/blogs/grumpyolddba/
Viewing 2 posts - 16 through 16 (of 16 total)
You must be logged in to reply to this topic. Login to reply