SpiNNaker2 tapes out

SpiNNaker2 tapes out

Based on the SpiNNaker infrastructure developed by Manchester University , the hybrid AI architecture created at TU Dresden, the adaptive body bias IP platform ABX by Racyics, and utilising GLoFo’s 22nm FD-SOI process,  SpiNNaker2 is a realtime neuromorphic AI processor with sub-millisecond latency for event-based systems.

“There is still a great deal to learn from biology if we are to realize the full potential of AI in the future,” says  ARM inventor Professor Steve Furber of Manchester University (pictured), “SpiNNaker2 has been designed to bridge the gulf between realistic brain models and AI so that each may increasingly be informed by the other”.

 SpiNNaker2 is designed to scale up to a 70,000 chip cloud system while maintaining strict real time operation.

It is claimed that the processor will have a disruptive impact on AI applications which are beyond current hardware because of their triple demands on low latency, high throughput, and high energy efficiency. 

Author: Yoyokuo