Re: after dbm_stats sample_size same as num_rows

Hello Jeffrey,
yes, this works as designed and is based on NDV algorithm.

However this is not Oracle 19c specific - this implementation was alread there in 11g ;-)

For more information please check these blog posts ...

• https://blogs.oracle.com/optimizer/how-does-autosamplesize-work-in-oracle-database-11g
• https://blogs.oracle.com/optimizer/how-does-auto_sample_size-work-in-oracle-database-12c

... or check this doc if you wanna go really deep and low-level ...

• https://jonathanlewis.files.wordpress.com/2011/12/one-pass-distinct-sampling.pdf

Best Regards
Stefan Koehler

Independent Oracle performance consultant and researcher Website: http://www.soocs.de
Twitter: _at_OracleSK

> Beckstrom, Jeffrey <jbeckstrom_at_gcrta.org> hat am 02.03.2021 19:49 geschrieben:
>
> Trying dbms_stats with auto sample_size per below code in 19.7 versus specifying of a sample size. Sample size is same as num_rows for all tables. Is this as expected?
>
> dbms_stats.gather_table_stats(
> ownname => obj_rec.owner,
> tabname => obj_rec.table_name,
> method_opt =>
> 'for all indexed columns size 1 ',
> estimate_percent => dbms_stats.auto_sample_size,
> cascade => true,
> block_sample => true,
> degree => dbms_stats.auto_degree);
>
> Jeffrey Beckstrom
> Greater Cleveland Regional Transit Authority
> 1240 W. 6th Street
> Cleveland, Ohio 44113

--
http://www.freelists.org/webpage/oracle-l