Vector Collection Schema

Data artifactKnowledge & DataKnowledge & Dataarc:VectorCollectionSchema

An explicit schema for a vector store collection declaring the fixed embedding dimensionality plus stored chunk text, source identifier, JSON metadata and an auto-generated primary key.

Responsibility. Fixes the structure and embedding dimension of stored vector records.

Also known as: Collection schema

Direct neighbourhood (hover for relationship types)

Relationships

configures structural

Chunk Schema Validator Ch6.3B
Vector Index Store abstract Ch6.3B

constrains control

Vector Batch Ingestor Ch6.3B

Design guidance

MUST declare the embedding dimension to match the embedding model output; it cannot change after collection creation.
SHOULD store chunk text alongside vectors to avoid joins with a separate document store at query time.
SHOULD store a source-document identifier per chunk so neighbouring chunks can be fetched for additional context.

Classification

Patterns: Co-located text and vectorsChunk-to-source linkage for progressive context expansionAuto-generated primary keys
Risks mitigated: Insertion failures from embedding dimension mismatchDuplicate-ID failures during concurrent batch loadsJoin latency between vector results and a separate document store

Sources

Ch6.3B: T. Nguyen, "ETL Worked Example - Load Phase & Pipeline Integration," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 6.3B. ISBN: 9798244538229.