RE: wait events

John,

Based on my experience, this ratio can indicate that a session is using an index that needs a rebuild.

I have seen sessions with a ratio greater than 25% (and increasing) and giving shocking performance. After rebuilding the indexes in question the ratio for the session drops below 5% and the performance improves significantly.

We have a job that rebuilds such problematic indexes on a regular basis.

Steve.
-----Original Message-----
From: john.j.kanagaraj_at_shell.com.bn
[mailto:john.j.kanagaraj_at_shell.com.bn]
Sent: Thursday, May 18, 2000 12:44 PM
To: Multiple recipients of list ORACLE-L Subject: RE: wait events

>Check the ratio 'db file sequential reads' to 'table fetch by rowid'
for
>each session. If the ratio is GT 10% then you may have excessive
>fragmentation in an index. The solution is to rebuild the index.

>Steve.

Steve,

How do you compare a WAIT statistic, i.e. 'db file sequential reads' from V$SESSION_EVENT (which is a measure of number of waits/timeouts/time waited) to a EVENT count from V$SESSTAT i.e. 'table fetch by rowid'? In other words the wait statistics provides an indication of how many times a session waited, not how many times it occured... Do you see what I mean?

Not trying to beat down your argument, just trying to follow the thought. Was this from a document somewhere?

Cheers,
John Kanagaraj
Brunei Shell Petroleum
http://www.geocities.com/john_sharmila

God so loved the world that He didn't send a committee! (See John 3:16 for details)
** Opinions expressed here are solely mine and not necessarily those of Received on Thu May 18 2000 - 00:35:41 CDT