Comments on: I wish http://jonathanlewis.wordpress.com/2011/12/13/i-wish/ Just another Oracle weblog Mon, 20 May 2013 17:10:14 +0000 hourly 1 http://wordpress.com/ By: I Wish « Oracle Scratchpad http://jonathanlewis.wordpress.com/2011/12/13/i-wish/#comment-43228 Fri, 16 Dec 2011 18:32:35 +0000 http://jonathanlewis.wordpress.com/?p=7812#comment-43228 [...] couple of days ago I wrote about some things I’d like to see in index statistics, which means changing dbms_stats.gather_index_stats(). Here’s an idea for [...]

]]> By: Log Buffer #250, A Carnival of the Vanities for DBAs | The Pythian Blog http://jonathanlewis.wordpress.com/2011/12/13/i-wish/#comment-43208 Fri, 16 Dec 2011 06:02:03 +0000 http://jonathanlewis.wordpress.com/?p=7812#comment-43208 [...] Jonathan wishes that it would be nice efficiency feature if Oracle changed the code to use the new Approximate NDV mechanism for these counts. [...]

]]> By: Timur Akhmadeev http://jonathanlewis.wordpress.com/2011/12/13/i-wish/#comment-43143 Wed, 14 Dec 2011 07:07:38 +0000 http://jonathanlewis.wordpress.com/?p=7812#comment-43143

It would be nice efficiency feature if Oracle changed the code to use the new Approximate NDV mechanism for these counts.

Yes indeed it would be an easy win. One point though: Approximate NDV is very CPU intensive operation, so DEGREE should also be considered in this case. On the positive side DBMS_STATS in 11g with auto sample size uses something like 10% of index blocks via IFFS and produces almost 100% accurate statistics.

]]>