We've started experimenting with using BRIN indexes for some of our large tables, as a replacement for B-Tree on "natural" timestamp columns that seem to be a good case for BRIN. Is it correct that BRIN indexes don't support MIN/MAX operations ?
We see that a query like `SELECT max(timestamp_column) FROM table` does a sequential scan on the table instead of using the index.
We're on Postgres 10.23 at the moment if that's important.
