Re: Slow response to my query
Keith <keith@keithf4.com>
From: Keith <keith@keithf4.com>
To: Bzzzz <lazyvirus@gmx.com>
Cc: Goke Aruna <goksie@gmail.com>, Steven Pousty <steve.pousty@gmail.com>, pgsql-novice@lists.postgresql.org, Babatunde Adeyemi <barbietunnie@gmail.com>
Date: 2019-11-29T15:01:16Z
Lists: pgsql-novice
On Fri, Nov 29, 2019 at 9:56 AM Bzzzz <lazyvirus@gmx.com> wrote: > On Fri, 29 Nov 2019 15:38:53 +0100 > Goke Aruna <goksie@gmail.com> wrote: > > > Thanks Jean-Yves / Steven, > > am using V12. > > Depending on the way you mostly use your columns, you can also do the > opposite: store date/timestamp into "pieces" in different columns > (century, month, day, hours, minutes, seconds, milliseconds, time zone > shift) and reconstitute a whole date/ts into an auto-generated column. > > Sometimes, you have to break the rules, especially with a large number of > rows, because pre-calculation is way better than post in this case. > > Depending on your needs, you might also be interested into that: > https://severalnines.com/database-blog/guide-partitioning-data-postgresql > > BTW, don't leave us dry, share your progress and results *<;-) > > Jean-Yves > Please do not advise the use of inheritance/trigger based partitioning anymore, especially on PG12. Partitioning is now part of core and is much more easily managed that way https://www.postgresql.org/docs/12/ddl-partitioning.html