Crap, I spent a lot of time formatting that table, and now it looks
crappy due to the font!
RogBaker_at_gmail.com wrote:
> Oracle 8.1.7.4
> I am confused by the different namespaces in the Library Cache. I did a
> statspack report, and this is what I see:
>
> Library Cache Activity for DB: FSQT Instance: fsqt
> Snaps: 3 -4
> ->"Pct Misses" should be very low
>
> Get Pin Inva-
> Requ Pct Requ Pct Re lidat
> Namespace sts Miss ests Miss loads ions
> --------------- ----- ---- ----- ----- ------ -----
> BODY 4 0.0 4 0.0 0 0
> CLUSTER 0 0 0 0 0
> INDEX 0 0 0 0 0
> OBJECT 0 0 0 0 0
> PIPE 0 0 0 0 0
> SQL AREA 110 0.0 69,672 0.0 0 0
> TABLE/PROCEDURE 182 17.0 237 53.6 0 0
> TRIGGER 0 0 0 0
> --------------------------------------------
>
> As I understand it, the Library Cache is used to hold parsed SQL
> statements so that they can be reused without hard parsing to reduce
> CPU. I am concerned about the 17% and the 53.6% for the
> TABLE/PROCEDURE. I am not sure what this means. If you do a SELECT
> ENPNAME, DEPT FROM EMPLOYEEDATA WHERE EMPID=567, does this go into the
> SQL AREA or the TABLE/PROCEDURE AREA? The docs I have been reading only
> seem to list the namespaces and do not provide any further details or
> information. Does it look like we need a bigger library cache/shared
> pool? I believe we are using bind variables for the most part.
>
>
> Thanks,
Received on Fri Jun 03 2005 - 14:18:21 CDT
