|1:00 - 1:05||Opening|
|1:05 - 1:40||Architectural Considerations for the Internet of Things||Prabal Dutta
U of Michigan
|1:40 - 2:15||RoboBees: A convergence of body, brain, and colony on silicon||David Brooks
|2:15 - 2:50||Nonvolatile processor architecture exploration for energy harvesting in IoT applications||Yuan Xie
|2:50 - 3:10||Break and open discussion|
|3:10 - 3:55||Intelligent IoT - algorithmic & architectural lessons from the brain||Bipin Rajendran
New Jersey Institute of Technology
|3:55 - 4:30||Visual Cortex on Silicon||Chita Das
|4:30 - 5:05||Programming the Web of Things||Vijay Janapa Reddi
U of Texas at Austin
The rapid growth in the number of smart, connected devices—trillions expected within a decade or so—has been widely predicted by industry leaders and popularly labeled the Internet of Things. However, adding computation ("smarts") and communication ("connections") to everyday things is just the first step. To really flourish, today’s battery lifetime-limited and largely stove-piped solutions that address a narrow set of well-defined problems must morph into shared, distributed, modular, and adaptive designs that can be used to realize an interconnected Web of Things. The challenges, no doubt, start with energy and connectivity issues at the lowest layers of the system but quickly encompass the rest of the system stack, spanning embedded, gateway, and cloud computing platforms, and integrating many disparate technical underpinnings to realize real systems that can be deployed. In this talk, I will explore these challenges by drawing from my own research which seeks to give individuals, communities, and countries greater visibility into the physical processes that matter to them.
Prabal Dutta is a Morris Wellman Faculty Development Associate Professor of Electrical Engineering and Computer Science at the University of Michigan. He is a systems builder who creates new devices and integrates them into real-world systems that attack challenging societal problems. His work has yielded dozens of hardware and software systems, has won four best paper awards, has received several design awards, has been directly commercialized by a dozen companies and indirectly by many dozens more, and has been utilized by thousands of researchers and practitioners worldwide. His work has been recognized with an NSF CAREER award, an Alfred P. Sloan Research Fellowship, an Intel Early Career Award, and a Popular Science Magazine Brilliant Ten Award. He holds a B.S. in Electrical & Computer Engineering and an M.S. in Electrical Engineering, both from the Ohio State University, and a Ph.D. in Computer Science from UC Berkeley.
The Harvard Robobees Project is a large-scale multidisciplinary project that seeks to develop a swarm of autonomous, life-sized robotic insects. The driving application for small scale mechanical flying insects is to autonomously pollinate a field of crops, while other applications range from search-and-rescue to mobile sensing. Construction of these devices provide many challenges that span computer science, mechanical engineering, electrical engineering, and biological sciences. This talk will describe an accelerator-based computing architecture implemented in 40nm CMOS that provides the necessary energy/performance requirements for the Robobee Brain in a highly-integrated form factor.
Bio: David Brooks is the Haley Family Professor of Computer Science in the School of Engineering and Applied Sciences at Harvard University. Prior to joining Harvard, he was a research staff member at IBM T.J. Watson Research Center. Prof. Brooks received his BS in Electrical Engineering at the University of Southern California and MA and PhD degrees in Electrical Engineering at Princeton University. His research interests include resilient and power-efficient computer hardware and software design for high-performance and embedded systems. Prof. Brooks is a Fellow of the IEEE and has received several honors and awards including the 2012 ACM Maurice Wilkes Award, NSF CAREER award, IBM Faculty Partnership Award, and DARPA Young Faculty Award.
Energy harvesting has been widely investigated as a promising method of providing power for IoT applications. Such energy sources include solar energy, radio-frequency (RF) radiation, piezoelectricity, thermal gradients, etc. However, the power supplied by these sources is highly unreliable and dependent upon ambient environment factors. Hence, it is necessary to develop specialized systems that are tolerant to this power variation, and also capable of making forward progress on the computation tasks. In this talk, we discuss how to use NVM technology to design nonvolatile processor and explore the design space for a nonvolatile processor with different architectures, different input power sources, and policies for maximizing forward progress.
Yuan Xie is a professor at UC Santa Barbara (UCSB). He received Ph.D. from Princeton University, and then joined IBM Microelectronics as advisory engineer. From 2003 to 2014 he was with Penn State as Assistant/Associate/Full professor. He also took on-leave and worked with AMD between 2012-2013. His research interests include EDA/architecture/VLSI, and has published more than 200 papers in IEEE/ACM venues. He was elevated to IEEE Fellow for contributions in design automation and architecture for 3D ICs. More information can be found at: http://www.ece.ucsb.edu/~yuanxie
The human brain consists of approximately 100 billion neurons that communicate with each other through a network of 1000 trillion synapses. More impressively, all the cognitive, control and creative processes that the brain seamlessly execute are performed within a power budget of about 20 W. Clearly, IoT platforms of today pale in comparison with this engineering marvel of nature, both in terms of power eciency and computational intelligence. In this talk, I will introduce some of the basic aspects of computational `hardware' and `software' of the brain, and discuss some ideas for embedding cognitive algorithms and systems within IoT platforms.
Bio: Bipin Rajendran received the B.Tech degree from I.I.T. Kharagpur, in 2000, and the M.S and Ph.D degrees in Electrical Engineering from Stanford University, in 2003 and 2006, respectively. He was a Master Inventor and Research Sta Member at IBM T. J. Watson Research Center in New York during 2006-'12 and a faculty member in the Electrical Engineering Department at I.I.T. Bombay during 2012-'15. His research focuses on building algorithms and systems for brain-inspired computing. He has authored over 45 papers in peer-reviewed journals and conferences,and has been issued 50 U.S. patents. He will start as an Associate Professor in the Department of Electrical & Computer Engineering at New Jersey Institute of Technology in January 2016.
Our visual cortex is highly sophisticated in terms of processing dynamically changing scenes with minimal energy consumption. The objective of this multidisciplinary project is to explore the feasibility of designing an end-to-end visual cortex system on silicon so that it can be integrated with a mobile platform and subsequently in the IoT ecosystem. This talk covers an overall view of the complete design space covering the understanding of the human cognitive system, mapping it to vision algorithms, exploring the hardware design alternatives, and the human interface issues for three application domains – visually impaired, driver assistance and augmented reality. The talk will conclude with some architectural design challenges for the mobile/IoT domain.
Chita Das is a Distinguished Professor of Computer Science and Engineering at the Pennsylvania State University. His main areas of interest include CMPs and manycore architectures, GPUs, handhelds, performance evaluation, fault-tolerant computing, and Clouds/datacenters. In particular, he has worked extensively in the area of design and analysis of interconnection networks/on-chip interconnects.
The Internet of Things is about the melding of hardware and software together to create a smarter ecosystem. But the Internet by itself is simply a network that allows devices to talk to one another. It is the Web, and its host of technologies, that has the potential to enable useful end-user application services that can transform society. Thus, the Internet of Things is really about the Web of Things. Programming the Web of Things introduces new challenges and opportunities for research in the architecture community. The talk focuses specifically on edge IoT devices, and it presents a case study of a visual event-driven programming paradigm built on top of Web technologies. It describes how event-driven programming as a fundamental construct can enable low-power, high-throughput edge IoT architectures.
Vijay Janapa Reddi is an Assistant Professor in the Department of Electrical and Computer Engineering at The University of Texas at Austin. His research interests include system architecture and software design and implementation to address performance, power and reliability issues for mobile and high-performance systems. He is the recipient of the Intel Early Career Award, PLDI Most Influential Paper Award, and has received Best Paper and Top Picks awards in Computer Architecture. Beyond his research activities, Vijay is passionate about STEM education, involving computer science education starting at an early age. He created the Hands-On Computer Science (HaCS) course that teaches computer science to 6th and 7th graders in the Austin Independent School District (AISD) using Arduinos and Raspberry Pis. Vijay received his Ph.D. in Computer Science from Harvard University.
Internet-of-Things (IoT) is expected to dominate the electronics world of 2020s and beyond. IoTs are not sensor nodes, they are expected to process sensor data and communicate useful information. The need for relatively complex functionality but at a resource-constrained environment is likely to require innovation in computing micro-architecture. Ultra-low-power design is a must, non-conventional computing may be required, security remains a challenge. The goal of this workshop to start a discussion on what roles micro-architecture can play in shaping the world of IoT.
Original contributions are invited in the areas of:
The selected papers will be presented as a talk or a poster.