I have a scenario where I need to add a unique constraint to an existing hypertable in TimescaleDB, which currently has compressed data. The process I am following now involves decompressing all the data before adding the unique constraint and then compressing it again.
Given the large volume of data, this approach is causing us to run out of memory during decompression.
TimescaleDB 2.14 | db<>fiddle (dbfiddle.uk)
Is there an alternative method to add the unique constraint without decompressing the data, considering the memory limitations we’re facing?
3 posts - 2 participants