Santa Clara Valley IEEE Communications Society at the Inaugural Fog World Congress
IEEE Communications Society of Silicon Valley (comsocscv.org) is pleased to present a special event at the inaugural
Fog World Congress being held October 30 - November 1, 2017 in Santa Clara, CA
Wednesday, November 1: 3:30 - 5:00 P.M. (open to all attendees and general public)
Innovations at the Edge: ASIC, Compute & Networking Perspectives.
Moderators: MP Divakar / Sameer Herlekar
Talk #1: IC Designs for IoT Edge Devices – Doing More with Less
By Fergus Casey, R&D director (ARC Processors), Synopsys
Abstract: The Internet-Of-Things (IoT) market is causing the focus of IC design flows to shift from increasing performance to reducing overall energy consumption in order to meet the demands of battery-operated devices.
As a consequence, Power-efficient processors and SoC architectures are required at the Edge to accommodate the scaling performance requirements from always-on sensing operations to the more demanding voice and face recognition applications, while living within the overall power profile required by IoT Edge ICs. This presentation will provide an overview of features and techniques to increase energy efficiency using Synopsys ARC processors.
Fergus Casey is an R&D director for ARC Processors at Synopsys, with engineering responsibility for the ARC EM, ARC 600 and ARC 700 families. He joined ARC International in 2003 as a processor verification engineer and has worked in various roles within the ARC processor group as part of ARC International and through the acquisitions by Virage Logic and later Synopsys. Prior to joining ARC, Fergus worked in a number of fabless semi and IP companies and start-ups in Ireland and UK, including Toucan Technologies, PMC-Sierra and Icera. Fergus holds a bachelor's degree in electrical engineering from University College Cork, Ireland.
Talk#2: Distributed Architectures for IoT Event Processing and HPE’s Approach to Edge Computing
By Lin Nease, Chief Technologist for HP Enterprise (IoT)
Abstract: The vast majority (90%+) of compute workloads today execute their logic in “people time” – i.e. CRUD logic triggered by human action, as part of some business process or consumer action. This is why the standard compute model has evolved so much: from proprietary systems thru open systems thru multi-tier thru virtualized thru cloud thru microservices via REST. Bottom line: CPU cycles and network bandwidth have not been nearly as important as software productivity and infrastructure cost. The next wave of computing, however – “physics time” – demands distributed topologies and more real-time event processing. This session will discuss the nuances of “physics time” topologies, and what HPE is doing to enable them (particularly with the edge).
Lin Nease is the Chief Technologist for HP Enterprise’s IoT activities. In this role, He is responsible for setting strategy, building a technology plan, and driving innovation with key enterprise customers/partners of HP. He also established HP’s membership in the Industrial Internet Consortium and has led multiple activities in working groups and testbeds for the Consortium. In his 25+ years with HP, he has been a Chief Technologist and Director of strategy for multiple business units, including HP’s Business Critical Servers and Networking businesses, has been a Chief Technologist and General Manager for multiple global accounts (GE, UPS), driven multiple M&A activities and cross-business initiatives, led numerous successful commercial products, including the industry’s first blade solution and HP’s long-lived Superdome platform, and holds several patents in software-defined networking. Nease received his BS in Computer Science at Arizona State University and his MBA from California State University Sacramento. He also served as a computer operator in the United States Air Force.
Talk#3: Connectivity at the Edge: Closing the loop
By Alfred Gomes, CEO, Axiomware
Abstract: For most applications, the characteristics of the underlying data-flows should drive the choice between cloud and fog computing. Unfortunately, this choice is not as simple as data-flows depend on two exponentially growing technologies: Computing and communication. This talk explores critical architecture choices that enable deployment of these technologies while guarding against continuous paradigm shifts and technology obsolescence.
Alfred Gomes leads Axiomware's efforts to provide seamless connectivity to our customers operating at the intersection of IoT, communication technologies, sensor innovations and big data. Alfred's passion for creating high-value, high-impact products is enabled by a rich experience in hardware, software and system design. Before Axiomware, he held multiple roles at Texas Instruments and National Semiconductor, with a focus on sensor signal path products. He was instrumental in the development of many capacitive, inductive and environmental sensing products. Alfred holds a PhD from Georgia Tech in the area mixed-signal semiconductor engineering and has multiple patents and publications in this field.