June 2024 Autonomous Database

Included are some notable Oracle AI Vector Search updates with the Autonomous Database, June 2024 release.

Feature	Description
HNSW Index Support in RAC Environments	It is possible to use Hierarchical Navigable Small World (HNSW) indexes with SELECT statements in RAC environments. For more information, see HNSW Support in RAC Environments.
Sizing the Vector Pool with ADB-S	With Autonomous Database Serverless (ADB-S) services, the vector pool dynamically grows and shrinks. It cannot be explicitly set. For more information, see Size the Vector Pool.
SQL Quick Start with Vector Generator	A SQL scenario is provided to get you started using Oracle AI Vector Search. The quick start includes a PL/SQL program that creates a vector generator, offering a simple alternative to using a vector embedding model. For the full tutorial, see SQL Quick Start Using a FLOAT32 Vector Generator.
Similarity Search and Index Creation Syntax	The APPROX and APPROXIMATE keywords are now optional. If omitted while connected to an ADB-S instance, an approximate search using a vector index is attempted if one exists. For more information about approximate search with indexes, see Approximate Search Using HNSW and Approximate Search Using IVF. The NEIGHBOR keyword in the ORGANIZATION clause of CREATE VECTOR INDEX is now optional. For the full syntax to create HNSW and IVF indexes, see Hierarchical Navigable Small World Index Syntax and Parameters and Inverted File Flat Index Syntax and Parameters.