| From: | Alvaro Herrera <alvherre(at)2ndquadrant(dot)com> |
|---|---|
| To: | Josh Berkus <josh(at)berkus(dot)org> |
| Cc: | PostgreSQL Advocacy <pgsql-advocacy(at)postgresql(dot)org> |
| Subject: | Re: Correlated columns & firsts |
| Date: | 2017-07-26 17:46:18 |
| Message-ID: | 20170726174618.hbs5o2gvqpmi2hqj@alvherre.pgsql |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-advocacy |
Josh Berkus wrote:
> Folks,
>
> I had the impression that our new multicolumn stats feature was a first
> in the industry. However, a PDXPUG member said that SQL Server has
> something like it for the last couple versions.
>
> Anyone know if it's the same thing or not?
A quick search led me to
https://dba.stackexchange.com/questions/84747/statistics-are-multicolumn-histograms-possible
which contains a pointer to
https://blogs.msdn.microsoft.com/ianjo/2005/11/10/create-multi-column-statistics/
then you can see the MS docs at
https://docs.microsoft.com/en-us/sql/t-sql/statements/create-statistics-transact-sql
https://docs.microsoft.com/en-us/sql/relational-databases/statistics/statistics
From what I can tell, they only store the correlation between the
columns, so it's simpler than what we do.
--
Álvaro Herrera https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Daniel Westermann | 2017-07-26 17:52:59 | Re: Correlated columns & firsts |
| Previous Message | Peter Geoghegan | 2017-07-26 17:42:49 | Re: Correlated columns & firsts |