Re: Index help

From: M.Godlewski <mcgodlewski_at_yahoo.com>
Date: Mon, 26 Jul 2004 07:31:58 -0700 (PDT)
Message-ID: <20040726143158.4803.qmail@web42008.mail.yahoo.com>

Data distribution.  

IDA3A5     count(IDA3A5)
4104                           44
4107                         156
4110                             1
4111                             1
4137                             1
46847                      4905

Daniel Fink <Daniel.Fink_at_Sun.COM> wrote: Assuming a uniform distribution, an index would probably make sense. If the data is skewed (1 value is in 80% of the rows, etc.), an index may not make sense. If the CBO uses bind peeking at parse time, you could end up with a poor execution plan. For example, the first call uses the 80% value in the predicate, so Oracle decides to use a FTS. Subsequent executions use the other values or even a value that does not exist. These executions are also going to use an FTS, when an index lookup is probably faster.

Daniel

M.Godlewski wrote:
> 6 distinct values for ida3a5

---

Please see the official ORACLE-L FAQ: http://www.orafaq.com

---

To unsubscribe send email to: oracle-l-request_at_freelists.org put 'unsubscribe' in the subject line.

--
Archives are at http://www.freelists.org/archives/oracle-l/
FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------

		
---------------------------------
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!

----------------------------------------------------------------
Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
To unsubscribe send email to:  oracle-l-request_at_freelists.org
put 'unsubscribe' in the subject line.
--
Archives are at http://www.freelists.org/archives/oracle-l/
FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------

Received on Mon Jul 26 2004 - 09:29:03 CDT