Use timescale's continuous aggregates for optimizing queries

[camallen]

It is desirable to avoid recalculating query data each time, timescale offer a pre-computed 'view' through, https://docs.timescale.com/latest/using-timescaledb/continuous-aggregates

Use these continuous aggregate 'views' to precompute expensive time bucket operations for the hypertable (i.e. select time bucket window of data for aggregates over the hypertable).

Note as of timescale 1.4 they do not support filter clauses but they intend to in the future.
https://docs.timescale.com/latest/using-timescaledb/continuous-aggregates#future-work

Also they intend to allow the 'view' to keep the underlying data and drop the raw chunks values to allow the data to be rolled up for long term storage.