Re: Reading/Interpreting 11g Statspack reports

Try to post your query and it's own plan. Are you able to use Sql advisors and maybe force a different plan? Are u really sure u'll not be able to talk with your programmers to change the query?

Consider also that sometimes an FTS maybe more efficient than using indexes.

Ste

--

http://www.stefanocislaghi.eu

------Original Message------

From: Janine Sisk
Sender: oracle-l-bounce_at_freelists.org
To: oracle-l L
ReplyTo: janine_at_furfly.net
Subject: Reading/Interpreting 11g Statspack reports Sent: Mar 9, 2010 06:26

Any favorite resources out there for deciphering these reports? For 8i/9i I used the excellent breakdown in Tom Kyte's "expert one on one" book, but it is failing me now; too much has changed. i tried searching the official docs but all I could find was that Statspack has been replaced by AWR, but I'm not licensed to use that. Funny how all roads lead to higher license fees! :) Google has not been fruitful either.

In case it matters, here is what I'm trying to do: I've been using JMeter to benchmark an old Linux server running 8i and an Amazon Linux instance running 11g. Same application (a website written in Java) and same data. For a relatively lightweight page the Amazon instance is much faster, as it should be, but for a database-intensive page it's slower. This made me think that I need to do some tuning/configuring.

I have run explain plan on the worst performing query and see a distressing number of "table access full" scans, but am trying to restrain myself from working on that, partly because it's not my code or data (I'm just the sysadmin) and partly because it's the same on the 8i system so as ugly as that is, it "shouldn't" be the problem.

I am open to suggestions as to better tools besides Statspack, keeping in mind that I'm a one-woman shop and can't afford Enterprise software fees and I also don't run Windows unless I absolutely have to. :)

thanks,

janine

--

http://www.freelists.org/webpage/oracle-l

Sent by BlackBerry®†Ûiÿü0ÁúÞzX¬¶Ê+ƒün– {ú+iÉ^ Received on Tue Mar 09 2010 - 03:15:32 CST