
Low Latency Routing in Space
About this video
A short visualization from HotNets researchers of routing and latency over SpaceX's proposed Starlink constellation, based on its FCC filings. It shows orbital shells, link topology and how inter-satellite routes could beat terrestrial fiber on long distances.
Why it appears in the archive
It is the go-to visual for how Starlink's constellation and laser links could cut global latency, linked in many satellite internet threads.
AI-generated summary and labels from the source dataset; verify important details against the video. Source & credits.
From the Hacker News discussion
This video [0] really puts into perspective what SpaceX is trying to accomplish. Incredible stuff if they pull it off! [0] [VIDEO]
Read original comment ↗There was recently a video posted theorizing this here: [VIDEO] Very interesting watch. Based off the paper: [LINK]
Read original comment ↗I really recommend watching this simulation of how the Starlink Constellation would massively cut global internet latency: [VIDEO]
Read original comment ↗This video goes a bit into the probable topology of the links and gives a good look at how the eventual constellation would look. Eventually in the full constellation there will be 4 easy links to the birds in front, behind, left and right, those 4 will stay in largely the same relative position until they reach the highest latitude of their orbit. Those alone will make an ok link with a bit of a zig zag. The tough ones are those ~180 degrees out of phase those will be crossing roughly perpendicular and at a really high speed. [VIDEO]
Read original comment ↗This constellation should have lower latency over long distance than is even theoretically possible using terrestrial fiber, because the speed of light in space is substantially better than the speed of light in fiber. [0] Even over short distance the latency (which the "Legal Narrative" pdf quotes as 15ms) is negligible for almost all applications. [0] paper [LINK] / video [VIDEO]
Read original comment ↗There were recently two simulations done on possible latencies based on the FCC filings. You can find the video simulations in the links at the bottom of this post. You can see international links that have latencies lower than the existing internet by a significant margin. Unfortunately neither of those simulations take into account that the first iteration has been changed to only use radio and no laser links. [0]: [VIDEO] [1]: [LINK]
Read original comment ↗First, check out the video explaining the orbits: [VIDEO] Doing some quick math, the shell in which they'll be orbiting has a surface area of 373,367,808km. With ~7000 satellites, an even distribution would leave them 533,382 km between satellites. All the satellites have station keeping thrusters to keep them in LEO with drag. Keeping them from running into each other is probably pretty simple, in comparison to the other challenges they'll need to solve.
Read original comment ↗The first video mentions at 6m55s[1] that as part of the third phase they're deploying satellites in high inclination orbits, among other reasons to provide coverage for Alaska, since this is an FCC requirement. I wonder if it's exclusively because of that reason, and if so that the reason international customers will enjoy coverage at high northern and southern latitudes is because of the accident of history that the US ended up acquiring Alaska, and that there's some FCC regulation that says you need to provide coverage for all 50 states. 1. [VIDEO]
Read original comment ↗