With the release of MariaDB 11.6, the MariaDB Foundation has announced the public preview of Vector search for the open-source fork of the MySQL engine. Database experts and open-source advocates see vector support as an opportunity for MariaDB to lead the MySQL ecosystem, especially since Oracle reserves most new features for its enterprise editions only.
According to the announcement, a feature complete MariaDB 11.6 Vector Edition is expected to be released within a few months, with vector functionality fully available on MariaDB 11.7 and beyond. Unlike MySQL 9.0 which currently offers a new vector type without Indexes, MariaDB does not yet provide a dedicated data type but has introduced dedicated indexes and the functions VEC_ToText and VEC_FromText, a new option for storing vectors.
The vector search implementation is based on the industry standard HNSW algorithm and the new MariaDB Vector, a storage engine for LLMs demoed at FOSDEM earlier this year, is on its way to becoming part of the standard MariaDB server. Kaj Arnö, CEO of the MariaDB Foundation, writes:
What makes us different from the Vector support in MySQL Server – called Heatwave – is that we have vector search as Open Source. In fact, unlike MySQL, we still lack a separate Vector data type, but we will add it before the full release.
Peter Zaitsev, founder at Percona and open source advocate, comments:
This project is obviously in the early stages but shows one important case where MariaDB is beating Oracle MySQL in things that matter to modern developers - Vector Search. I'm curious what Oracle's reaction will be. Will it compel Oracle to make Vector Search actually useful in MySQL Community edition or does Oracle not care about their Open Source version being relevant for modern developers anymore?
Arnö adds:
We are looking at helping not just the MariaDB Server users but also the MySQL Server users to get easy Open Source access to Vector functionality. We’ve always made the point that migrating from MySQL Server to MariaDB Server is dead easy, and getting access to Vector functionality is a great reason to get going.
The implementation of vector search results from collaborative work by employees of MariaDB plc, the MariaDB Foundation, and other contributors, particularly from AWS, one of the most recent supporters of MariaDB.
With generative AI workloads getting more popular, MariaDB is not the only relational database working on support of vector search. As previously reported on InfoQ, Google and AWS have introduced this option in their managed RDBMS, and MySQL 9.0 is the first Oracle release working on vector search support. Furthermore, PlanetScale announced last year its intention to fork MySQL to add vector search.
While many developers appreciate the different alternatives for vector search, user nhatnv comments on Reddit:
Unless you are dealing with a scale of billions, I don't see the need to have many options for vector DBs.
In the article "How Fast Is MariaDB Vector?", Sergei Golubchik, chief architect at MariaDB plc, compares the new feature with other vector databases using the ann-benchmarks suite.
Being an alpha release, no full binary packages are currently available; however, a container image supporting vector search with CPU optimizations enabled is now available at Quay.io.