Topic
tsvector columns with GIN indexes as an alternative to a separate Elasticsearch or Meilisearch cluster.
turbopuffer says v3 cut hot and cold full-text p90 latency by up to 80% against v2 on 10-million-document namespaces. The figure is from the vendor's own nightly benchmark, and the company does not expect a gain for cold vector queries.
Reality
MariaDB rebuilt a 320 MB table for a plain ADD COLUMN and blocked a write for over 30 seconds when a FULLTEXT key was present, a dev.to test found. An identical table without the index took 60 milliseconds.
Reality
The extension puts BM25 ranking behind an ordinary Postgres index, so a newly committed row is searchable without a trip to an Elasticsearch cluster. PlanetScale's own benchmark reports 199 queries per second against ParadeDB's 7.9.
Reality
Aurora Serverless v2 idles at about $43 a month at its 0.5-ACU floor, close to three times the whole $15 EC2 box being retired, so the read path became a SQLite file in S3 and the writes went to DynamoDB.
Reality
A side project swapped RabbitMQ, Redis and Elasticsearch for SKIP LOCKED, an unlogged table and a GIN index. The queue arithmetic shows where that choice stops working.
Reality