hey all,
I've gotten my data importing quite nicely now and its working great. I'd like to now add a layer of multi-tenancy on my data cube: my facts table contains a lot of data for a lot of users and I'd like to segment it on a userid.
I'm thinking of setting up schemas in postgres (where I store my cube). this would work for the querying, but not so much for the ETL, since I import using the COPY command.
The kneejerk solution would be to have some kind of post-import script that would take my facts table and partition it into the schemas, but I'm not a fan of the solution.
I feel like there must be a pattern for solving this nicely.
I would appreciate all thoughts here :)