Top all-time · #5327
Jonathan Katz
Personal blog · jkatz05.com/postdataprogramming
Jonathan Katz's personal blog at jkatz05.com covers data and programming. Since 2023, 14 different Hacker News readers have linked it 19 times, most often “Scalar and Binary Quantization for Pgvector Vector Search and Storage”. No feed was found, so read it on the site itself.
Too few sampled comments (4) to say how often HN recommends it.?
Why HN links it
Jonathan Katz's posts benchmark pgvector and Postgres vector search, including HNSW index performance and compression. Commenters cite them for benchmark data on vector indexing.
Summarised by Claude from a sample of 4 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Most-linked posts?
- Scalar and Binary Quantization for Pgvector Vector Search and Storage4 people · 4 mentions · first linked 2024-04-09
- Distributed Queries for Pgvector3 people · 3 mentions · first linked 2024-03-18
- An early look at HNSW performance with pgvector2 people · 3 mentions · first linked 2023-08-10
- Will PostgreSQL ever change its license?2 people · 2 mentions · first linked 2024-04-19
- Pgvector 0.5.0: HNSW indexing support2 people · 2 mentions · first linked 2023-08-29
- The 150x PGVector Speedup2 people · 2 mentions · first linked 2024-05-01
What Hacker News says?
Have a look at https://jkatz05.com/post/postgres/pgvector-performance-150x-... fp16 for indices seems to give a nice size compression without reducing quality with HNSW.
Comment linking /post/postgres/pgvector-performance-150x-speedup/
on “I’m writing a new vector search SQLite Extension”Submitted Thoughts on PostgreSQL in 2024 as a story · 124 points
HN discussionI would like to throw our project in the ring that solves this problem: https://github.com/tjmlabs/ColiVara 1. Uses half-vecs, so you cut down everything by half with no recall loss 2. Uses token pooling with hierarchial clustering at 3, so, you further cut down things by 2/3rd with <1% loss 3. Everything is on Postgres and pgvector, so you can do all the Postgres stuff and decrease corpus size by document metadata…
Comment linking /post/postgres/pgvector-scalar-binary-quantization/
on “VectorChord: Store 400k Vectors for $1 in PostgreSQL”Submitted Hybrid Search with PostgreSQL and Pgvector as a story · 35 points
HN discussionJonathan did just that in his prior blog post: https://jkatz05.com/post/postgres/pgvector-hnsw-performance/ pgvector_hnsw outperforms pg_embedding using a query per sec / recall measure across various common embedding widths and a range of dataset sizes.
Comment linking /post/postgres/pgvector-hnsw-performance/
on “Pgvector 0.5.0 Feature Highlights and how tos”Submitted The 150x PGVector Speedup as a story · 35 points
HN discussion> HNSW indices it should perform well on large amounts of vectorized data as well (though we haven't tested this yet) here are a few benchmarks from this week's commit: https://jkatz05.com/post/postgres/pgvector-hnsw-performance/
Comment linking /post/postgres/pgvector-hnsw-performance/
on “You just do not need a specialized vector database”Submitted An early look at HNSW performance with pgvector as a story · 11 points
HN discussionSubmitted Distributed Queries for Pgvector as a story · 5 points
HN discussionSubmitted Pgvector 0.5.0: HNSW indexing support as a story · 5 points
HN discussionSubmitted Will PostgreSQL ever change its license? as a story · 4 points
HN discussionSubmitted Scalar and Binary Quantization for Pgvector Vector Search and Storage as a story · 4 points
HN discussionSubmitted Will PostgreSQL ever change its license? as a story · 2 points
HN discussionSubmitted Distributed Queries for Pgvector as a story · 2 points
HN discussionSubmitted Pgvector 0.5.0 Feature Highlights and how tos as a story · 2 points
HN discussionSubmitted Scalar and Binary Quantization for Pgvector Vector Search and Storage (2024) as a story · 1 points
HN discussionSubmitted Scalar and binary quantization for pgvector vector search and storage as a story · 1 points
HN discussionSubmitted Distributed Queries for Pgvector as a story · 1 points
HN discussionSubmitted Vectors Are the New JSON in PostgreSQL as a story · 1 points
HN discussion
