Comments on: Oracle 11g: Incremental Global Statistics On Partitioned Tables

By: Andre van Winssen

Andre van Winssen — Wed, 02 Dec 2009 10:01:12 +0000

Hi,
I just located a patch for 11.2.0.1 (using oracle’s advanced classic patch search on https://updates.oracle.com/Orion/SavedSearches/switch_to_advanced.. :-)
8686932: HISTOGRAMS NOT PRESERVED AFTER ENABLING DBMS_STATS INCREMENTAL STATISTICS
appearantly histograms should be preserved after enabling incremental statistics but without this patch they aren’t.

Andre

By: //TheOracles בלוג המומחים של אורקל » ארכיון » איסוף סטטיסטיקות על טבלאות ב-11g

Thu, 30 Jul 2009 16:23:31 +0000

[...] למי שמעונין לקרוא קצת יותר על זוג היכולות האחרונות, כולל מדידת הביצועים, תציצו כאן או כאן. [...]

By: Kerry Osborne’s Oracle Blog » Blog Archive I’ll Gladly Pay You Tuesday for a Hamburger Today - Kerry Osborne’s Oracle Blog

Thu, 21 May 2009 00:33:40 +0000

[...] info). I should also mention that Greg Rahn has several good posts on 11g Statistics Enhancements here, here, and [...]

By: amit poddar

amit poddar — Mon, 21 Jul 2008 00:36:48 +0000

Hi, We have been using new granularity APPROX_GLOBAL AND PARTITION which comes with patch 6526370. It aggregates all the global statistics (including histograms) from partition level statistics except NDV. Following is the description from dbmsstat.sql

--         'APPROX_GLOBAL AND PARTITION' - This option is similar to
--         'GLOBAL AND PARTITION'. But the global statistics are aggregated
--         from partition level statistics. It will aggregate all statistics
--         except number of distinct values for columns and number of distinct
--         keys of indexes.
--         The existing histograms of the columns at the table level
--         are also aggregated.The global statistics are  gathered
--         (i.e., going back to GLOBAL AND PARTITION behaviour)
--         if partname argument is null or if the aggregation cannot be done
--         e.g., statistics of one of the partitions is missing.
--         This option is useful when you collect statistics for a new
--         partition added into a range partitioned table (for example,
--         a table  partitioned by month).
--         The new data in the partition makes the global statistics stale
--         (especially the min/max values of the partitioning column).
--         This stale global statistics may cause suboptimal plans.
--         In this scenario, users can collect statistics for the newly added
--         partition with 'APPROX_GLOBAL AND PARTITION' option so that the
--         global statistics will reflect the newly added range.
--         This option will take less time than 'GLOBAL AND PARTITION' option
--         since the global statistics are aggregated from underlying
--         partition level statistics.
--         Note that, if you are using APPROX_GLOBAL AND PARTITION,
--         you still  need to collect global statistics (with granularity =
--         'GLOBAL' option)
--         when there is substantial amount of change at the table level.
--         For example you added 10% more data to the table.
--         This is needed to get the correct number of distinct values/keys
--         statistic at table level.

By: Log Buffer #106: A Carnival of the Vanities for DBAs

Log Buffer #106: A Carnival of the Vanities for DBAs — Fri, 18 Jul 2008 21:04:42 +0000

[...] Oracle performance guru (and fellow Wisconsin native) Greg Rahn makes a great case for using incremental global statistics on partitioned tables. [...]

By: Greg Rahn

Greg Rahn — Wed, 16 Jul 2008 20:54:03 +0000

Histograms are (can be) gathered at the partition level and then aggregated to form a global/table level histogram. As long as the same method_opt command is used from the start, it is possible. Only in the scenario where the method_opt changes triggering a new histogram collection on a partition, is it a requirement to look at data from the entire table. This is because not all partitions will have histograms (only the ones where stats have been gathered since the method_opt change). In this case there is a scan of the whole table done using a small sample to create the global histogram.

Is histogram data stored in the synopsis: No.
Can histograms on partitions be aggregated to form a global histogram when they exist on each partition: Yes.

Hopefully that makes it a bit more clear.

By: Christian Antognini

Christian Antognini — Wed, 16 Jul 2008 20:36:08 +0000

Hi Greg

IMO it is important to point out that incremental gathering is not available for histograms.

Cheers,
Chris