Post

PlanetScale vectors is now GA! We have made vectors a first-class part of MySQL - focusing on robustness to make sure vector support is as solid as every other data type MySQL supports. pscale.link/vectors

Writes and queries work like a normal RDBMS. Store your vector data in a column in your existing database, build an index with an ALTER or CREATE VECTOR INDEX statement, or write SELECT statements with JOIN and WHERE clauses like any other data.

One of the biggest differentiators of PlanetScale vector support is the ability to use indexes larger than RAM. We are the first to incorporate an index based on SPANN (Space-Partitioned Approximate Nearest Neighbors) into an RDBMS.

SPANN is a hybrid index design that combines the best aspects of partition-based and tree-based indexes. In SPANN, vectors are assigned to small partitions, which are stored in hidden InnoDB tables.

Only the tree and a small, fixed number of relevant partitions need to be in memory when building and querying the index, allowing efficient operation with vector data up to 6x bigger than available memory.

Vectors especially shines when running on Metal. Using vector indexes on PlanetScale Metal ensures that loading vector partitions from InnoDB to answer queries will be as fast as possible. pscale.link/metal