Researchers at Trinity College Dublin have created a new light-based technology on a tiny chip. This advance could make data centres, which power cloud computing, artificial intelligence, and global internet services, faster and more efficient. The work appears in Nature Communications.
The researchers used microscopic ring-shaped devices known as microresonators - tiny loops that trap and circulate light - to produce extremely stable light signals. These signals create optical frequency combs, which act like optical rulers. They generate many evenly spaced colours of light that allow very precise measurements and data transmission. The researchers also produced a new kind of stable light pulse called a hyperparametric soliton. This pulse lets the comb signals appear at different colours from the main laser powering the device. The demonstration happened in the wavelength range used for high-speed links inside large data centres.
Chip-scale optical innovation for data centre efficiency
The new approach simplifies high-speed optical communications that move large amounts of data through fibres by sending many colours of light at once, a method called wavelength-division multiplexing. Optical frequency combs can replace groups of separate lasers, making systems simpler, more stable, and more efficient. Data centres already use large amounts of electricity. Growing demand from artificial intelligence increases these needs further. Improvements in efficiency could help reduce electricity consumption and support lower carbon emissions.
The researchers noted that the new optical source will interest experts in optical communications and precise measurements. The work involved theorists at the University of Bath and specialists in microresonator fabrication at the Swiss Federal Institute of Technology Lausanne. It also connects to an Irish company developing related laser technology. Researchers expect this to form the basis for stronger future developments in optical networks.
Overall, the technology offers a practical step toward next-generation data infrastructure that handles rising computing demands with better energy performance.