Re:How We Built Uber’s Highest Query per Second Service Usin...
by rosshemsley on Tuesday March 01, 2016 @07:07PM
(#11207219)
Yes, but the O(N) is generally only for pathological inputs.
Geo data is generally nicely distributed. R-tree (or similar) is the correct solution here.
Re:How We Built Uber’s Highest Query per Second Service Usin...
by urschrei on Tuesday March 01, 2016 @06:49PM
(#11207124)
Was just looking for an implementation last week. Is it public?
Re:How We Built Uber’s Highest Query per Second Service Usin...
by openasocket on Tuesday March 01, 2016 @07:05PM
(#11207212)
Not yet, but it will probably be open sourced sometime soon.
Source: just spent two months implementing R-trees in Rust