Space servers

 One of the benefits of Elon Musks' Starlink space internet is that it provides a lower-latency connection between points on earth. It does this because the speed of light in a vacuum is faster than the speed of light in fibre and so even though the beams have to travel slightly further* they travel that distance faster.

Although this is great for long-distance zoom chats the real financial benefit of this is to high-frequency traders who make a living from the small arbitrage opportunities that exist between geographically separate exchanges.

To that end, I conjecture that not long after the public release of Starlink and its proven utility to high-frequency traders, those trading firms will start to deploy their algorithms to servers in orbit.

Why? To lower the latency. If the algorithm doesn’t need to send the signal to space before it executes then it can receive the signals from both exchanges in L/2 and after deciding to trade send the trading request at L/2 leading to a blisteringly quick L+computation time to execute. 

This led to a larger and larger number of satellites in what, realistically, is a limited number of orbits. Limited, not because there isn’t enough physical room but, because much like running out of airspace for commercial aircraft it’s about safely being separated.

* Even that might not be the case given the signal might have to be grouped into local exchanges first.


Popular posts from this blog

An exploration in number systems

Structural engineering with cardboard