I'm dealing with a very large partitioned table (several millions of inserts per day) and i solve this by running `vacuum analyze` nightly which somehow solved 99% our read issues
Prior to that, we tried various indexing strategies, de-normalizing some of the data, but ultimately i found that weird plan inconsistency to be solved by the vacuum job
Prior to that, we tried various indexing strategies, de-normalizing some of the data, but ultimately i found that weird plan inconsistency to be solved by the vacuum job