Comments on: Partition Count http://jonathanlewis.wordpress.com/2007/01/02/partition-count/ Just another Oracle weblog Fri, 17 May 2013 13:58:17 +0000 hourly 1 http://wordpress.com/ By: Hash Partitions « Oracle Scratchpad http://jonathanlewis.wordpress.com/2007/01/02/partition-count/#comment-7700 Mon, 07 May 2007 21:07:05 +0000 http://jonathanlewis.wordpress.com/2007/01/02/partition-count/#comment-7700 [...] Infrastructure — Jonathan Lewis @ 9:06 pm UTC May 7,2007 A question recently appeared on an article I wrote about partitioning a few months ago: We are planning to create 8 HASH partitions. Looking only at PERFORMANCE would be [...]

]]> By: Karteek http://jonathanlewis.wordpress.com/2007/01/02/partition-count/#comment-7596 Sun, 06 May 2007 11:07:14 +0000 http://jonathanlewis.wordpress.com/2007/01/02/partition-count/#comment-7596 on Oracle 9i R2

We are planning to create 8 HASH partitions. Looking only at PERFORMANCE would be there be any improvements, if we go for 16 or 32 partitions (maintainance and availability is not a problem in our case). There are only 2 indexes on our 350M table – one is LOCAL, another non-partitioned index.

Jon…any advice that you would give..?

- Karteek

]]> By: Jonathan Lewis http://jonathanlewis.wordpress.com/2007/01/02/partition-count/#comment-1122 Fri, 05 Jan 2007 09:29:36 +0000 http://jonathanlewis.wordpress.com/2007/01/02/partition-count/#comment-1122 Alexander, Oh yes I would!

]]> By: Alexander Fatkulin http://jonathanlewis.wordpress.com/2007/01/02/partition-count/#comment-1121 Fri, 05 Jan 2007 09:25:05 +0000 http://jonathanlewis.wordpress.com/2007/01/02/partition-count/#comment-1121 Some time ago we had table with more than 90 000 segments (range + list composite partitioning). You wouldn’t believe how much time it took to parse query for the first time…

]]> By: Jonathan Lewis http://jonathanlewis.wordpress.com/2007/01/02/partition-count/#comment-1060 Wed, 03 Jan 2007 08:29:33 +0000 http://jonathanlewis.wordpress.com/2007/01/02/partition-count/#comment-1060 Antonio, I think the closest thing you may find would be the scattering of “best practices” papers that Oracle’s CoE has published over the years on Metalink.

Point 2 appeared in 9i with an optional change in syntax appearing in 10g, but point 3 only appeared in 10g. There is a (probably small)price to pay for it.

The data dictionary views that reports partitions still display sequential partition numbers, which they derive by applying the analytic function rownumber() to the underlying data. So if you regularly query things like the dba_tab_partitions view, then the resource usage for some of your queries may jump significantly – especially if you have large numbers of partitions in your system.

]]> By: Antonio http://jonathanlewis.wordpress.com/2007/01/02/partition-count/#comment-1043 Tue, 02 Jan 2007 13:41:27 +0000 http://jonathanlewis.wordpress.com/2007/01/02/partition-count/#comment-1043 Mumble!

This post make me think!

Not on the statements that you say but on a question a bit more general:

are there Oracle’s patterns?

In other words: are there standard strategies developed on how optimizer works and how Oracle stores data?

Another: point 3 happed in Oracle 9? :-|

]]>