Relational Vector Extension Store

Data storeKnowledge & DataKnowledge & Dataarc:RelationalVectorExtensionStore

A general-purpose relational database extended with a vector type, distance functions and vector indexes, storing embeddings beside relational data under ACID transactions.

Responsibility. Stores vectors alongside relational records with transactional consistency.

Also known as: PostgreSQL vector extension

Variant of Vector Index Store abstract

When to choose. Choose when an organization already operates the relational database at scale, datasets are moderate (<1M vectors), and transactional consistency across relational and vector operations is required.

Direct neighbourhood (hover for relationship types)

Relationships

alternative to variability

Design guidance

MAY be chosen to reuse existing relational expertise, backup, monitoring and security tooling without new operational dependencies.
SHOULD be complemented by, or migrated to, a specialized vector store when latency degrades beyond ~1 million vectors.

Quantitative guidance

As stated by the sources; verify before use.

Reasonable below ~1 million vectors; latencies increase substantially beyond (Ch6.2A).

Classification

Patterns: IVFFlat and HNSW indexesACID transactions spanning relational and vector dataMigrate hot vectors to a specialized store while keeping relational data in place
Technologies: pgvectorPostgreSQL
Quality attributes: Reliability (ISO/IEC 25010 | NIST AI RMF: valid and reliable)Maintainability (ISO/IEC 25010)

Sources

Ch6.2A: T. Nguyen, "Vector Database Selection," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 6.2A. ISBN: 9798244538229.