Commercial time-series data historians often use swinging door compression as this reduces storage required, enhances retrieval efficiency, minimizes noise, simplifies analysis, and enhances scalability.
This is very useful for both IIoT and Plant Historian use-cases.
Do people see the value in adding this feature?
How would we go about adding this as a feature?
Hi Steve, that’s a great question!
Timescale doesn’t have any swinging door compression but you can achieve very similar results with downsampling functions.
Timescaledb compression is all lossless and I think LTTB and ASAP Smooth are very closer to what you expect.
Swinging Door Compression is more about reducing data transmission and storage at the sensor or device level, often used in real-time systems to minimize the data that needs to be sent over a network. TimescaleDB Compression, on the other hand, is used at the database level to optimize storage and query performance in server or cloud environments.
Anyway, feel free to suggest it](Sign in to GitHub · GitHub), more users can upvote and we can discover if this is useful from the community perspective.