● Covered by 1 source · 1 reportMedium impact1 neutral
Tiger Data Releases Open Source pg_vectorsearch for Scalable Vector Search in PostgreSQL
Key points
- pg_vectorsearch is an open-source PostgreSQL extension.
- Uses PRISM index for scalable vector similarity search.
- PRISM index offers faster builds and lower memory use.
- Supports existing pgvector columns.
New PostgreSQL Vector Search Extension
Tiger Data, the company behind TimescaleDB, has released pg_vectorsearch, an open-source PostgreSQL extension for vector similarity search. The extension is currently in an alpha stage. It aims to provide more scalable and efficient vector search capabilities directly within PostgreSQL.
PRISM Index Architecture
The core of pg_vectorsearch is its PRISM index, which utilizes data partitioning and compression. This approach differs from the graph-based HNSW (Hierarchical Navigable Small World) index used by pgvector. PRISM groups similar vectors into partitions and stores them in a compressed format, densely packed into database pages. This design reduces memory requirements and index build times.
Performance and Efficiency Improvements
Tiger Data's benchmarks indicate significant performance gains. In tests with 100 million vectors, the PRISM index built in 1 hour and 40 minutes on a machine with 128 GB of memory. In contrast, HNSW took 42 hours on a machine with 512 GB of memory for the same dataset. The PRISM index also combines hierarchical clustering, compressed vectors, and RaBitQ quantization to optimize search and distance calculations.
Integration and Impact
pg_vectorsearch supports existing pgvector columns, allowing applications to perform vector searches without requiring a separate database. This integration simplifies infrastructure for handling large vector datasets directly within PostgreSQL, potentially reducing operational complexity and costs for developers working with vector embeddings.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
Primary sources
GitHub timescale/timescaledb GitHub timescale/rekall GitHub timescale/pg_vectorsearchReporting from
Tiger Data released pg_vectorsearch, an open-source PostgreSQL extension for vector similarity search, currently in alpha. The extension uses a PRISM index with data partitioning and compression, offering faster and more memory-efficient vector search compared to graph-based approaches like HNSW. This allows developers to handle large vector datasets directly within PostgreSQL without needing a separate database.