Hacker News Reader

To be precise, it's impossible to measure one-way latency between a pair of nodes without external information. But if you have a mesh of nodes, the story is different [1]. If you have N nodes and hence N unknown clock offsets, and if you have ping data from N(N-1) pairs, you can do better.

[1] https://www.usenix.org/conference/nsdi18/presentation/geng

View on HN

Well I think we need to distinguish between practical latency measurements on earth, where it's mostly a matter of rejecting dodgy data / outliers etc.., and detecting systematic latency bias that is theoretically impossible to detect. If you could get around this problem just by using a network of N nodes, then you could do stuff like measuring an absolute reference frame, thereby violating special relativity.

Reply on news.ycombinator.com