December 24, 2003 at 6:07 am
Taken from BOL as I could not put it in my own words 🙁
This is just a basic explaination...
When any SQL statement is executed in SQL Server 2000, the relational engine first looks through the procedure cache to verify that an existing execution plan for the same SQL statement exists. SQL Server 2000 reuses any existing plan it finds, saving the overhead of recompiling the SQL statement.
If no existing execution plan exists, SQL Server 2000 generates a new execution plan for the query. (this results in the time consumption when the procedure is run 1st time after SQL Server starts)
After an execution plan is generated, it stays in the procedure cache.
The age query plan is decremented, unless another user references the plan
HTH
Sachin
🙂
Regards,
Sachin Dedhia
December 24, 2003 at 7:30 am
quote:
I am running a sp on SQL2000 on Win2k machine with 128 MB ram.
It could be shortage of memory resource. It runs longer first time could be caused involving I/O activities to load all necessary data it needs to run. After that, all data is in the cache and it runs faster next time.
You should review the execution plan each time you feel the sp behaves stranger.
December 25, 2003 at 12:25 am
quote:
It could be shortage of memory resource
I have a sp that runs for approx 2 min on SQL2000 on Win2k machine with 128 MB RAM and approx 30 seconds with 256 MB RAM ![]()
But the repeated execution of the same sp on
SQL2000 on Win2k machine with 128 MB RAM does not makes comparable difference as compared to with 256 MB RAM.
Sachin
![]()
Regards,
Sachin Dedhia
January 7, 2004 at 7:13 am
thanks sachindedhiya fo ryour suggestions.
![]()
Viewing 4 posts - 1 through 5 (of 5 total)
You must be logged in to reply to this topic. Login to reply