
AI models are permeating ever more applications, expanding the industry’s appetite for computational power. The conventional response has been to pile on more GPUs and build ever-larger data centres in a seemingly endless race for energy and brute force.
The work was published in the journal National Science Review, and its corresponding authors include Shu Haowen and Wang Xingjun from Peking University.
The “joints” connecting these FPGAs are two custom-designed communication hardware components. The first is a silicon photonic transceiver chip running at 400 gigabits per second, responsible for converting electrical signals to optical and vice versa.

Don't Miss:
-
China university trademarks emblem, sparks speculation over potential Armani logo dispute
-
Europe recorded 10,000 excess deaths in late-June heatwave
-
Hong Kong Observatory may issue T1 signal between Monday night, Tuesday morning
-
North Korea’s Kim fires warning shot at own military with corruption purge
-
Restaurant pullouts from dog-friendly scheme ‘responsible’, 20 complaints recorded

Thailand in Parade of SEA Cyberscam Nations
BOOK REVIEW: City on Fire: A Novel of Hong Kong
Diane Thompson on Russell Vought’s Campaign to Abolish the CFPB