Knowledge & Data · Data artifact
IVF Index Configuration
Data artifactKnowledge & DataKnowledge & Dataarc:IVFIndexConfig
A vector index configuration that partitions vectors into k-means clusters (nlist) and searches exactly within only the most relevant clusters.
Responsibility. Configures cluster-partitioned approximate vector search.
Also known as: IVF_FLAT index, Inverted file index
Variant of Vector Index Build Configuration abstract
When to choose. Choose as a balanced default for mid-size collections needing good recall with manageable memory.
Relationships
alternative to variability
Design guidance
- SHOULD set the cluster count near the square root of the total vector count, capped around 1,024-4,096.
Quantitative guidance
As stated by the sources; verify before use.
- Reduces search space by ~90-95%, giving ~20x speedup with about 95% recall (nlist=1024, 10-50 clusters probed) (Ch6.3B).
- Suitable for collections under ~10 million vectors (Ch6.3A).
Sources
- Ch6.3A: T. Nguyen, "ETL Pipeline Fundamentals," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 6.3A. ISBN: 9798244538229.
- 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.