CCing to list to maintain context. On Sat, Feb 27, 2016 at 12:14 PM, Eric Mortensen <eric@xxxxxxxxxxx> wrote: > Thanks Francisco, I had not considered MVCC. If that is true, it would seem > to me that a GIN index would "always" be less efficient, as it potentially > would have to update every key's posting tree if a tuple moves, whereas two > btree indexes would only require modififying two trees. If you are that worried about performance, maybe your data will be best served by splitting those fields out of the jsonb or even splitting the table. As always, measure, identify bottleneck. Francisco Olarte. -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
