Poor man’s approach:
- use How to keep your recorder database size under control to determine the entities with largest footprint in your recorder database
- for each of the top, say, 5 entities: create a “filtered” version and use that in the recorder setup, e.g.
# To throttle a sensor down to 1 value per minute, the window_size should be set to "00:01”.
- platform: filter
name: "filtered smlreadertotalenergyconsumption"
entity_id: sensor.smlreadertotalenergyconsumption
filters:
- filter: time_throttle
window_size: "00:01"
Note that sensor.smlreadertotalenergyconsumption is the original sensor. In my record config, I’ve added sensor.filtered_smlreadertotalenergyconsumption.