Source: Wave Computing

Wave Computing CTO to Present Details of its Dataflow Technology at HOT CHIPS Symposium

CAMPBELL, Calif., Aug. 21, 2017 (GLOBE NEWSWIRE) -- Wave Computing, the Silicon Valley start-up that is revolutionizing machine learning with its roadmap of WaveFlow compute appliances, today announced its Chief Technology Officer (CTO), Dr. Chris Nicol, will be presenting at the HOT CHIPS Symposium on August 22, 2017, in Cupertino, CA.  Dr. Nicol’s presentation will share details of the company’s Coarse Grain Reconfigurable Array (CGRA) architecture, as well as its innovative Dataflow Processing Unit (DPU) chip, which recently taped out at the 16nm node. Wave Computing’s dataflow technology is being commercialized as a compute appliance for training and inferencing in data center environments. 

Title: A Course Grain Reconfigurable Gate Array for Training Deep Neural Networks
Date: Tuesday, August 22, 2017
Time: 10.45 am PT
Location: HOT CHIPS 2017 - Flint Center for the Performing Arts, Cupertino, CA

About Dr. Chris Nicol
Chris Nicol is the CTO and the lead architect of Wave Computing’s dataflow architecture. Prior to Wave, Dr. Nicol was CTO for Embedded Systems at NICTA, an Australian ICT R&D organization he helped to establish. He supported more than 20 R&D projects across five labs in hardware and software development for intelligent, secure, robust and power efficient embedded systems. Prior to NICTA, Dr. Nicol founded Bell Labs Research in Australia and Agere Systems in Australia. He received his PhD degree from the University of New South Wales and MBA from the Australian Graduate School of Management, and is an inventor on 21 U.S. patents. Dr. Nicol has served on the TPC of ISSCC twice, and is serving his second term on the TPC of IEEE International Symposium on Low Power Electronics and Design.

HOT CHIPS is one of the semiconductor industry’s leading conferences on high-performance microprocessors and related integrated circuits. The conference emphasis is on real products and realizable technology, with topics including embedded and reconfigurable processors, quantum computing, non structures, wireless chips, network / security processors, and advanced packaging technology.

About Wave Computing’s Dataflow Solution and Early Access Program
Wave Computing’s dataflow-based compute appliance is redefining machine learning by accelerating the performance and scalability of training and inferencing for deep and shallow neural networks. Initially optimized for the data center, each Wave compute appliance delivers up to 2.9 PetaOps per second of performance, more than 250,000 processing elements, and over 2 TB of high-speed memories.

To provide early access to qualified companies, Wave Computing has introduced an Early Access Program (EAP) whereby data scientists and developers will have cloud-based access to a Wave compute appliance prototype before official sales begin. Availability is limited, so interested companies should apply today at http://wavecomp.ai/early-access/

About Wave Computing
Wave Computing is the Silicon Valley start-up that is revolutionizing machine learning with its roadmap of WaveFlow compute appliances. The company’s solutions leverage its native dataflow technology to outperform machine learning products available today. Wave Computing, one of CIO Application Magazine’s Top 25 Artificial Intelligence Providers, is offering its solutions to customers globally.

Wave Computing is a trademark of Wave Computing, Inc. All other marks are the property of their respective companies.