Statistics catalogue

This is a list of all articles I’ve written that talk about statistics, with a date stamp and short note of the contents. I have also created a couple of sub-catalogues around major topics:

• Column Groups catalogue
• Histograms catalogue
• Partition Stats catalogue (other authors 2007 – 2012)

I am currently working backwards through my entire blog adding articles to catalogues as I go, so any individual catalogue may grow very slowly.

• 19c – better stats (April 2022): Oracle now uses approx_count_distinct() on the whole index in dbms_stats.gather_index_stats().
• system_stats() hint (Jan 2022): notes on a hint that allows you to set the “noworkload” system statistics in a hint on Exadata.
• Optimizer Tip (Sept 2021): reprint of a note for IOUG 2015 on column groups and the table_cached_blocks preference
• Stats preferences (Aug 2021): a simple query to report all the system and table preferences, default or explicitly set
• Using the num_index_keys() hint (Nov 2018): controlling the split between access and filter predicates used in the index for complex in-lists.
• Reporting times of stats operations (Sept 2018): dbms_stats functions to report details of stats gathering operations.
  • Older version of stats timing code (Sept 2016): relevant for versions without dba_operation_opstat_tasks.
  • Pre-12c stats timing code (Sept 2016): hack code to get an indication of time spent on a table gathering it stats.
• table_cached_blocks defect (July 2018): A table preference about indexes, but it isn’t used during create or rebuild of index.
  • table_cached_blocks and RAC (Nov 2015): if you set the table preference to adjust the index clustering_factor, should you allow for the effects of RAC?
  • table_cached_blocks (May 2013): At last, a “legal” way to adjust the clustering_factor to a realistic value.
• Optimising statspack (Oct 2016): selective stats collection for data dictionary / internal queries.
• v$lock efficiency (May 2013): the benefit of collecting stats on individual x$ structures
• Minimum Stats (Sept 2012): auto_sample_size on an index tends to mean 1,140 blocks

Other Authors

• Partition Stats (2007 – 2012) – a list of articles written by other people on the topic of staitistics on partitioned objects.
•