Phase-change memory enables energy-efficient brain-inspired computing

Abstract

As we embark on an era of artificial intelligence (AI) and cognitive computing, there is a pressing need for energy-efficient computing systems for highly data-centric AI related applications. It is becoming increasingly clear that to build efficient cognitive computers, we need to transition to novel architectures where memory and processing are better collocated. Computational memory is one such approach where the physical attributes and state dynamics of memory devices are exploited to perform certain computational tasks in place with very high areal and energy efficiency (Fig. 1) [1]. Phase-change memory (PCM) devices based on materials such as GeSbTe5 are appealing for computational memory owing to their multi-level storage capability and potential scalability. In this paper, we present a brief overview of our work towards enabling energy-efficient brain-inspired computing using PCM.