Thanks a lot Magnus for the reply.
I understand that jsonb_path_ops creates index on the whole path or on all
the attributes.
Does it takes more storage or reduces write performance in comparison to
normal gin index ?
On Tue, Mar 29, 2022 at 2:19 PM Magnus Hagander wrote:
>
>
> On Tue, Mar 29,
No matter what you do, you will not typically get the same level of
performance as normalized tables as you do not get statistics on the values
stored in your jsonb column. No knowledge of null fraction, number of
distinct values, or most common values and their frequencies.
On Tue, Mar 29, 2022 at 7:06 AM Saurav Sarkar
wrote:
> Hi All,
>
> We use JSONB /NoSQL functionality of PostgreSQL.
>
> One of the column "doc" in our table "Table1" is of type JSONB.
>
> Now the rows in this column "doc" can have different values with different
> schemas.
>
> For e.g values of d
