Linley Fall Processor Conference 2020

Join Imagination at The Linley Fall Processor Conference features with virtual presentations that run half days on October 20-22 and 27-29, 2020. Presentations will address processors and IP cores for AI applications, embedded, data center, automotive, and communications. Attendees will be able to view live-streamed presentations and interact with the speakers during Q&A, breakout sessions, or scheduled one-on-one meetings.

Linley offers free admission to qualified registrants making it easy to participate. The conference is intended for chip designers, system designers, equipment vendors, OEM/ODMs, service providers, press, and the financial community.

Title

Scalable Multicore Inference Engine for efficient compute at 100 TOPs

Abstract

With applications scaling from edge-based speech recognition to ADAS object detection and tracking, the need for higher-performance and scalable AI accelerators is imperative. The challenge is to minimize the impact on area and power while maximizing the utilization of memory bandwidth. This presentation will introduce a new multicore AI accelerator that is highly scalable so as to enable the most demanding markets. A novel tilling approach has been adapted to make the best of the multicore architecture.

Speaker

David Hough – Distinguished Systems Architect

David Hough is the Distinguished Systems Architect who leads on Imagination’s AI architecture. The award-winning NNA (Neural network Accelerator) has been designed by his team and features a series of architectural enhancements and improvements. The Series2NX NNA was the leading quant accelerator on AI-Benchmark in 2019 and the latest developments have pushed the envelope further, particularly in scaling multicore. He has been with Imagination for 14 years and studied at the University of York.