First rule of scaling Postgres = don't scale it, just scale your application properly. Don't build one big warehouse, make a lot of independent one that just sync time to time.
@@waats there are different monoliths. You can deploy your application per client/per request. So instead of having 10bil rows db you will have 20 x 50mil that easier to manage and not locking each other
man no fcking way, tetris in sql, mindblowing!
First rule of scaling Postgres = don't scale it, just scale your application properly. Don't build one big warehouse, make a lot of independent one that just sync time to time.
are you sure about this ?
to me it feel like poor man scaling .
In my humble experience monolith success lies in scaling the DB right ^^.
@@waats there are different monoliths. You can deploy your application per client/per request. So instead of having 10bil rows db you will have 20 x 50mil that easier to manage and not locking each other
Depends on the application. For a typical web app, that would work, but some data isn't amenable to segmenting in that manner.