ABSTRACT

The explosion of "big data" applications imposes severe challenges of data processing speed and scalability on traditional computer systems. The performance of traditional Von Neumann machines is greatly hindered by the increasing performance gap between CPU and memory, motivating the active research on new or alternative computing architectures. By imitating brain's naturally massive parallel architecture with closely coupled memory and computing as well as the unique analog domain operations, neuromorphic computing systems are anticipated to deliver superior speed for applications in image recognition and natural language understanding.

The objective of this research is to establish the fundamental framework and design methodology for NeoNexus -- the next-generation information processing system inspired by human neocortex. It integrates neuromorphic computing accelerators with conventional computing resources by leveraging large scale inference-based data processing and computing acceleration technique atop memristor crossbar arrays. The computation and data exchange will be carefully coordinated and supported by the innovative interconnect architecture, i.e., a hierarchical network-on-chip (NoC). The software-hardware co-design platform will be developed to address the various design challenges. The project will help computer architecture and high-performance computing communities to overcome the ever-increasing technical challenges of traditional architectures and accelerate the fusion between conventional computing technology and cognitive computing model. It will also promote the applications of artificial intelligence technology advances in modern computer architectures and motivate the inventions at both software and hardware levels. Undergraduate and graduate students involved in this research will be trained for the next-generation semiconductor industry workforce.

Please report errors in award information by writing to: awardsearch@nsf.gov.