To slow, or not to slow? New science in sub-second networks

TL;DR

This paper explores the scientific implications of intentionally introducing microsecond delays in ultrafast networks, highlighting the need for research to inform policy and improve understanding of system stability across various complex networks.

Contribution

It emphasizes the importance of scientific research on microsecond delays in ultrafast networks and advocates for access to exchange data to better understand system-wide impacts.

Findings

Microsecond delays can significantly affect network stability.

Understanding delays is crucial for safety in autonomous systems.

Research can inform policy decisions on network timing.

Abstract

What happens when you slow down part of an ultrafast network that is operating quicker than the blink of an eye, e.g. electronic exchange network, navigational systems in driverless vehicles, or even neuronal processes in the brain? This question just adopted immediate commercial, legal and political importance following U.S. financial regulators' decision to allow a new network node to intentionally introduce delays of microseconds. Though similar requests are set to follow, there is still no scientific understanding available to policymakers of the likely system-wide impact of such delays. Giving academic researchers access to (so far prohibitively expensive) microsecond exchange data would help rectify this situation. As a by-product, the lessons learned would deepen understanding of instabilities across myriad other networks, e.g. impact of millisecond delays on brain function and safety of driverless vehicle navigation systems beyond human response times.