I have a dataware house on my hands that works well but am trying to optimize.
We have a year dimension wich simply is a view to sequence
create view dim_year as
select `minidwh`.`seq_2012_to_2030`.`seq` AS `TDYEAR` from `minidwh`.`seq_2012_to_2030`
As month dimension we want to keep it simple and have a table
CREATE TABLE `dim_month` (
`TDMONTH` int(11) NOT NULL,
`month_de` varchar(10) NOT NULL
);
containing the months wich can be used in a outer join fasion to create a time dimension snow flake.
What would be the optimal storage engine for this? To remove overhead currently it is in CSV as it should small and cache all the time the dwh front end is using it.
Would this be optimal in a MariaDB/MySQL situation? Is there a better way to have a (multi lingual) month dimension?