4 ms·
1024 per vector x 41M vectors
by jbellis 2y ago
1024 per vector x 41M vectors
- hot_gril 2y ago1024-dim vectors would fit into pgvector in Postgres, which can do cosine similarity indexing and doesn't require everything to fit into memory. Wonder how the performance of that would compare to this.
- jbellis 2y agoIt's been a while since I read the source to pgvector but at the time it was a straightforward HNSW implementation that implicitly assumes your index fits in page cache. Once that's not true, your search performance will fall off a cliff. Which in turn means your insert performance also hits a wall since each new vector requires a search. I haven't seen any news that indicates this has changed, but by all means give it a try!
- hot_gril 2y agoThanks, I didn't know that. Last time I was dealing with these kinds of problems, pgvector didn't exist yet.