Settings

Theme

Binary Vector Search at 350GB/S Using ARM Neon

topk.io

7 points by Equiet 5 months ago · 2 comments

Reader

MarekDlugos 5 months ago

re: optimization for 1024b vectors — do you pad shorter ones, or fallback to a more general kernel?

  • marekgalovic 5 months ago

    We do a projection of the original vectors so that it matches one of our optimized kernel. This generally gives us better recall vs. simple padding since all bits are utilized.

Keyboard Shortcuts

j
Next item
k
Previous item
o / Enter
Open selected item
?
Show this help
Esc
Close modal / clear selection