Comment by arunmu
5 days ago
Didn't these guys also write pgvectorscale which seems to be unmaintained now ? Why a new index instead of adding onto or improving pgvectorscale?
5 days ago
Didn't these guys also write pgvectorscale which seems to be unmaintained now ? Why a new index instead of adding onto or improving pgvectorscale?
(pgvectorscale author here)
We address that directly in the blog post: https://www.tigerdata.com/blog/announcing-pg_vectorsearch-sc...
Summary: We weren't happy with the unpredictable recall performance of pgvectorscale (and couldn't find a way to fix that in the literature) and decided that moving away from graph-algorithms to partition-based ones was a better long-term approach.
Thanks. I wish you had gone into more details on why recall was bad and what scale etc so that current users could also identify their issues. Also, it seems the new index is memory based and does not scale on machines with lower memory by being built on disk data structure ?