Re: Increased storage size of jsonb in pg15

More information needed:

1) Table definition.

\d "DataRecords"
                Partitioned table "public.DataRecords"
   Column   |            Type             | Collation | Nullable |   Default  
------------+-----------------------------+-----------+----------+-------------
 filename   | character varying(255)      |           | not null |
 timestamp  | timestamp without time zone |           | not null |
 data       | jsonb                       |           | not null | '{}'::jsonb
 created_at | timestamp with time zone    |           | not null |
 updated_at | timestamp with time zone    |           | not null |
Partition key: RANGE ("timestamp")
Indexes:
    "DataRecords_pkey" PRIMARY KEY, btree (filename, "timestamp")
    "DataRecords_filename_updated_at_idx" btree (filename, updated_at)
Foreign-key constraints:
    "DataRecords_filename_fkey" FOREIGN KEY (filename) REFERENCES "DataFiles"(filename) ON UPDATE CASCADE ON DELETE CASCADE
Number of partitions: 97 (Use \d+ to list them.)

2) An example of reported size for the 14.? and 15.5 cases.

  Since upgrading from 14.8 to 15.5, the jsonb data that was previously written in 14.8 is reporting a smaller size than the same hourly data written after the upgrade (upgrade indicated in yellow):

  

file	hourly_timestamp	filename_bytes	timestamp_bytes	data_filesize	created_at_bytes	updated_at_bytes
datafile.dat	2023-10-19 12:00:00	23	8	1682	8	8
datafile.dat	2023-10-19 13:00:00	23	8	1687	8	8
datafile.dat	2023-10-19 14:00:00	23	8	1685	8	8
datafile.dat	2023-10-19 15:00:00	23	8	1668	8	8
datafile.dat	2023-10-19 16:00:00	23	8	2155	8	8
datafile.dat	2023-10-19 17:00:00	23	8	2178	8	8
datafile.dat	2023-10-19 18:00:00	23	8	2199	8	8
datafile.dat	2023-10-19 19:00:00	23	8	2187	8	8
datafile.dat	2023-10-19 20:00:00	23	8	2180	8	8
datafile.dat	2023-10-19 21:00:00	23	8	2176	8	8
datafile.dat	2023-10-19 22:00:00	23	8	2053	8	8
datafile.dat	2023-10-19 23:00:00	23	8	2043	8	8

3) Differences in the database attributes between the two instances.

  No differences.

4) Was this done on same machine? If not how are the machines different?

  Same AWS RDS Postgres instance.

5) How the upgrade was done?