Designing temporal networks that synchronize under resource constraints
Synchronization is critical for the function of many distributed systems—whether it's computers or power grids or neuronal populations—but doing it using the least amount of energy and resources possible can be a daunting task.
In a paper published in Nature Communications in June 2021, incoming SFI Postdoctoral Fellow Yuanzhao Zhang and former SFI external faculty member Steve Strogatz report using temporal network models to show that allowing connection patterns to change over time makes it possible to synchronize a system more efficiently.
"This was a fun project started by accident," says Zhang. "I was researching circadian clocks and came across an interesting paper about the energy cost of synchronizing them. It piqued my curiosity, so I wanted to figure out the best way to synchronize a generic networked system using the least amount of resources."
The researchers' temporal network design is "open loop," so it's versatile and expected to work for a wide range of systems.
More information: Yuanzhao Zhang et al, Designing temporal networks that synchronize under resource constraints, Nature Communications (2021). DOI: 10.1038/s41467-021-23446-9