Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
![]() |
![]() |
Home -> Community -> Usenet -> c.d.o.server -> Re: Query tuning
rajesh_at_solutionsoftware.com wrote in message news:<aa8abed4.0304101937.289ba7db_at_posting.google.com>...
OK, now we are talking.
> Here is the explain plan. Actually, I didnot know what statistics we
> use to analyze a query. So I couldnot furnish any further information.
> I am using ORACLE. 8.1.7.2.1. Please instruct if i need to furnish
> anything more. I donot know anything about tkprof either.
> SELECT STATEMENT
> SORT UNIQUE
> CONCATENATION
> MERGE JOIN OUTER
> MERGE JOIN OUTER
> MERGE JOIN
> SORT JOIN
> HASH JOIN
> INDEX FAST FULL SCAN
> IND_POLICYCUSTOMER1
> HASH JOIN
> TABLE ACCESS FULL
> MTOPTION
> HASH JOIN
> TABLE ACCESS FULL
> MTOPTION
> NESTED LOOPS
> HASH JOIN
> TABLE ACCESS FULL
> AGENT
> INDEX FAST FULL SCAN
> IND_AGENTPOLICY1
> TABLE ACCESS BY INDEX ROWID
> POLICY
> INDEX UNIQUE SCAN
> IND_POLICYID
> SORT JOIN
> TABLE ACCESS FULL
> CUSTOMER
> SORT JOIN
> TABLE ACCESS FULL
> CONTACTNO
> SORT JOIN
> TABLE ACCESS FULL
> ADDRESS
> MERGE JOIN OUTER
> MERGE JOIN OUTER
> MERGE JOIN
> SORT JOIN
> HASH JOIN
> INDEX FAST FULL SCAN
> IND_POLICYCUSTOMER1
> HASH JOIN
> TABLE ACCESS FULL
> MTOPTION
> HASH JOIN
> TABLE ACCESS FULL
> MTOPTION
> HASH JOIN
> TABLE ACCESS FULL
> AGENT
> NESTED LOOPS
> INDEX FAST FULL SCAN
> IND_AGENTPOLICY1
> TABLE ACCESS BY INDEX ROWID
> POLICY
> INDEX UNIQUE SCAN
> IND_POLICYID
> SORT JOIN
> TABLE ACCESS FULL
> CUSTOMER
> SORT JOIN
> TABLE ACCESS FULL
> CONTACTNO
> SORT JOIN
> TABLE ACCESS FULL
> ADDRESS
>
> "Mikito Harakiri" <mikharakiri_at_ywho.com> wrote in message news:<sjDka.9$Hg1.132_at_news.oracle.com>...
> > Tuning strategy might be different depending what oracle version you are in.
> > If you are on 9.2, then the easiest way to tune is to capture execution
> > statistics from V$SQL_PLAN_STATISTICS_ALL. Pay attention to the values in
> > the following columns:
> >
> > LAST_OUTPUT_ROWS
> > LAST_STARTS
> > LAST_CR_BUFFER_GETS
> >
> > Sometimes it is quite easy to spot performance hot spots. For example you
> > may see huge number of LAST_STARTS originating from some join node in the
> > rowsource tree, then it becomes obvious that the join method must be changed
> > from NL to HJ or SMJ. As a first step why don't you capture statistics and
> > let the group know it in order to suggest some directions for query tuning?
> >
> > <rajesh_at_solutionsoftware.com> wrote in message
> > news:aa8abed4.0304072132.65cddb37_at_posting.google.com...
> > > Here is a query that I am executing in my application and it takes
> > > around 10 minutes to execute. Absolutely no idea how to proceed with
> > > fine tuning.
Received on Fri Apr 11 2003 - 11:15:08 CDT
![]() |
![]() |