Have a question about other people’s experience.
Once I found this integration fell in love with it and spent days measuring various devices around my house!!
Recently I noticed my backup’s started taking forever and then I blew the limit on my Google Drive where the back up’s sync to. My database has ballooned over the past 3 months, whilst I’ve got the headroom on my hardware was curious.
Ran a quick query on the database and identified the biggest offenders were the sensors from PowerCalc (which when you think about it makes sense):
SELECT
COUNT(*) AS cnt,
COUNT(*) * 100 / (SELECT COUNT(*) FROM states) AS cnt_pct,
SUM(LENGTH(attributes)) AS bytes,
SUM(LENGTH(attributes)) * 100 / (SELECT SUM(LENGTH(attributes)) FROM states) AS bytes_pct,
entity_id
FROM states
GROUP BY entity_id
ORDER BY cnt DESC
source: How to keep your recorder database size under control
So I’ve been trying to get my head around what I want this data for. I use the current power in Lovelace next to the individual device – so I don’t need this in the recorder and could exclude it? I have my device group data in the energy dashboard, again I don’t think I need this in the recorder to source it there?
I was about to start a bit of testing around this, but any thoughts appreciated….
The second point of pondering is long term statistics, and I can’t find the answer – if I excluded these from the recorder would I still build the LTS data?
Of course I can reduce the duration of data retention to mitigate this, there are feature requests kicking around to allow this to be defined by entity which could help. As I say, just some thoughts and any feedback or similar experiences would be welcome