We are planning to use Dremio (GKE) to connect to a Postgres database. Plan is to speed up long analytics queries by using data reflections (raw / aggregated). I want to understand the best practices and if they are documented somewhere. The queries that we plan to run
- have joins spanning across multiple tables.
- have few filter conditions on columns have very high cardinality.
- have a lot of aggregations like COUNT, MIN,MAX,AVG,SUM etc.
- make use of windows functions