Depends what you are doing with those 100M rows. We have a database with about that many rows and we need to do sums of a few hundred thousand rows, as well as

Depends what you are doing with those 100M rows. We have a database with about that many rows and we need to do sums of a few hundred thousand rows, as well as bucket those rows into daily buckets and graph them in real time. With just regular Postgres these operations are pretty slow (> ~5 seconds), but with Timescale continuous aggregates and hypertables they run in milliseconds.