Dear all,
Of late, I have been involved in tuning the Performance for a
particular program with my project.
1) When you execute a particular query, say if it takes, 1 mins, the second run
will always take less time, like 10 msecs etc, the reason is that the second
run, benefits from the parsing done on the first run of the query.
One way to overcome, this is to use "alter system flush shared_pool" which is
normally "supposed" to clear the parsing in shared_pool, but that's where my
problem is, it sometimes does and most times doesn't, so I can't really say,
how mush I have tuned or is it tuned at all.
Any solutions ???????
2) The entire program must have only one commit, 'cos of that it takes a
longtime, 'cos of the huge amts of information's that need to be maintained in
the rollback segments.
Pragma autonomous_transaction commits Or backup tables are not an option.
Any methods, to enhance the speed to the program with rollback segment Or
methods to ease the load on the rollback segments????
Cheers
Kart
------------------------------------------------------------------------------------------------------------------------------------
Lot of failures in life are those who didn't realise how close they were to
success when they gave up !!
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/ocpfree/
<*> To unsubscribe from this group, send an email to:
[EMAIL PROTECTED]
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
---------------------------(end of broadcast)---------------------------
TIP 4: Don't 'kill -9' the postmaster
