Re: Reading Execution Plan
From: Hugo Kornelis (hugo_at_pe_NO_rFact.in_SPAM_fo)
Date: 12/23/04
- Next message: Bob Barrows [MVP]: "Re: Select * where 1 = 1"
- Previous message: Hugo Kornelis: "Re: "disallowed construct" on indexed view"
- In reply to: C TO: "Reading Execution Plan"
- Messages sorted by: [ date ] [ thread ]
Date: Thu, 23 Dec 2004 23:16:14 +0100
On Thu, 23 Dec 2004 09:41:01 -0800, C TO wrote:
>Hello World,
>
>May I ask how to reason the following:
>
>I have two queries which return exactly the same results (and 220 rows). I
>place the two queries and run a execution plan behind them. The one that
>takes 10 seconds by itself now with "Query cost (relative to the batch)"
>92.44%, while the other that takes 20 seconds with a "Query cost (relative to
>the batch)" only 7.56%.
>
>Based on this info, what kind of actions I should take?
>
>Thanks.
Hi C TO,
Did you clear out the cache between the executions? Run the following
commands before each query if you want to get meaningfull comparisons:
CHECKPOINT
DBCC DROPCLEANBUFFERS
DBCC FREEPROCCACHE
Best, Hugo
-- (Remove _NO_ and _SPAM_ to get my e-mail address)
- Next message: Bob Barrows [MVP]: "Re: Select * where 1 = 1"
- Previous message: Hugo Kornelis: "Re: "disallowed construct" on indexed view"
- In reply to: C TO: "Reading Execution Plan"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
