Postgres GIN Indexes for JSONB Containment Queries

A GIN index on a jsonb column makes @> containment queries fast by indexing every key/value pair inside the document, instead of scanning rows.

Without an index, WHERE data @> '{"status": "active"}' on a large jsonb column forces a sequential scan. Adding CREATE INDEX ON t USING gin (data) lets Postgres use the index for containment (@>), existence (?), and a few other jsonb operators.

Trade-offs:

• Write overhead: every insert/update touching the jsonb column updates the GIN index entries for each key.
• Index size can be large for documents with many keys — jsonb_path_ops is a smaller, faster variant if you only need @>.

Reference: PostgreSQL docs — GIN indexes