News for hackers. Stuff that matters.
Re:WyHLL: The most accurate 3-bits HyperLogLog by yorwba on Sunday March 14, 2021 @12:49PM (#26456730)
Is there a companion paper somewhere explaining how it works? I tried to read the comments in the source but didn't really understand much. (Might be related to the fact that I never knew much about vanilla HyperLogLog to begin with.)
Re:WyHLL: The most accurate 3-bits HyperLogLog by Manozco on Sunday March 14, 2021 @02:24PM (#26457485)
This article by antirez helped me understand HyperLogLog a couple years ago http://antirez.com/news/75
Re:WyHLL: The most accurate 3-bits HyperLogLog by ithkuil on Sunday March 14, 2021 @07:11PM (#26460460)
Thank you for the article, nice read.

Just a small nitpick, for the craic:

> Moreover, while accessing the registers, we need to compute the sum of pow(2,-register) which involves floating point math. > [...] > * The floating point computation was modified in order to allow for multiple operations to be performed in parallel when possible. This was just a matter of adding parens. Floating point math is not commutative, but in this case there was no loss of precision.

Floating point addition is commutative. It is not associative though.

Re:WyHLL: The most accurate 3-bits HyperLogLog by rahimiali on Sunday March 14, 2021 @04:58PM (#26459098)
based on the comments it appears to rely on this paper: https://arxiv.org/abs/1702.01284
Re:WyHLL: The most accurate 3-bits HyperLogLog by jalk on Sunday March 14, 2021 @01:06PM (#26456846)
Re:WyHLL: The most accurate 3-bits HyperLogLog by pwnguin on Sunday March 14, 2021 @01:17PM (#26456931)
But what about the actual software question? What makes this different than redis hyper log log? Or statsite