April 10, 2006 at 11:10 am
I've encountered several articles lately regarding the PREPARE...EXECUTE statements in combination with parameter markers improving SQL performance. I have only seen this method type of programming used once...while analyzing the profiler output on an SQL server receiving calls from one of our third-party compiled applications.
For all other apps, I have typically seen statements follow the standard ANSI SQL and T-SQL used for Dynamic, static, and ad-hoc situations.
So, can someone site some examples where as a DBA I should be using methods similar to the PREPARE...EXECUTE syntax...or pushing developers to use it? I've talked with several other programmers and DBA's and they've never used it either.
Is this typically a style of programming used by software shops that are distributing C based apps, etc...as opposed to web-based apps and apps that execute the bulk of their DB access through stored procs?
Ryan
April 10, 2006 at 1:42 pm
I found some information on this that primarily cleared up my question. It seems that because of the efficiency of plan reuse as of SQL2000, the PREPARE/EXECUTE model serves only to reduce network traffic...and ONLY reduces network traffic when three or more iterations of the same statement are used consecutively (or something close to that).
I would still be interested to here from anyone that is using this (in a non-proffessional software environment...that is, you aren't selling software) and how you are using it.
RH
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply