We have not observed any effects on relational table performance. Basically, TimescaleDB code in on-path for each parsed query, but if the query does not speci

We have not observed any effects on relational table performance. Basically, TimescaleDB code in on-path for each parsed query, but if the query does not specify a hypertable, we just let the query "fall through" to Postgres' standard query planning/execution. But you do point out a very common use case and strength of TimescaleDB: you can easily JOIN your time-series data (hypertables) with existing relational tables.