GitHub user yezhizi added a comment to the discussion: TimeSeries Proposal

Apologies, one point I seem to have overlooked in the proposal: `size` in 
metadata(or `totalSamples` in `TS.INFO`) explicitly represents the count of 
non-expired samples, similar to `firstTimestamp`. Therefore, it appears that 
merely updating the "size" when "sealing" is still insufficient.

> I'm considering the idea that leave 1 or 2 chunks "appending", and keep 
> `size` and other for the "likely sealed" chunks. So this will (1) only update 
> "size" when sealing (2) keep these info fast, without querying N chunks

GitHub link: 
https://github.com/apache/kvrocks/discussions/3044#discussioncomment-13727838

----
This is an automatically sent email for [email protected].
To unsubscribe, please send an email to: [email protected]

Reply via email to