Abstract: Cameras are everywhere - cities worldwide have deployed millions of cameras for planning and security purposes; industries ranging from manufacturing to health to IT heavily dependent on them for their day to day operation. These cameras record images 24x7x365, mostly storing them for possible analysis at a later time. The time lag between capturing and analysis is a limitation of the current technology and cost. I believe real-time video analytics of live and stored streams is compelling for many important reasons and also is a perfect application of edge computing. Unfortunately existing state-of-the-art video analytics systems are expensive, insufficient, and they often require manual intervention. Large-scale automated video analytics is a grand challenge for the research community and for those of us who work on big-data cloud-scale systems. Privacy regulations, bandwidth constraints and latency naturally lead us to design and develop systems where video is analyzed across both edge and cloud clusters. In this talk I will describe our hybrid edge-cloud video analytics infrastructure and a smart-city traffic system that we are deploying in the City of Bellevue in Washington, USA. I will discuss how we are aggressively moving towards commoditization and democratization of video analytics.
Bio: Victor Bahl is a Distinguished Scientist and the Director of Mobility & Networking in Microsoft Research. He is a member of the global leadership team overseeing MSR's world-wide operations. In these roles he works on matters related to the health and vibrant functioning of our domestic and international Labs. and advises Microsoft’s CEO and senior leadership team on long-term vision and strategy related to networked systems, mobile computing, wireless systems, cloud computing, and datacenter networking. He heads a high-powered group that executes on this vision through research, technology transfers to product groups, industry partnerships, and associated policy engagement with governments and research institutes around the world. Dr. Bahl has published over 125 scientific papers, authored over 140 patents, and won numerous technical and leadership awards incl. a test-of-time award, three best paper awards, two awards from the United States FCC, distinguished service and lifetime technical achievement awards from ACM SIGMOBILE, distinguished alumni award and a IEEE outstanding leadership award. Over his career he has developed seminal technologies including streaming video analytics (2015), white space networking (2010), edge-based cloud computing (2009), mesh networking (2005), multi-radio wireless systems (2001), Wi-Fi hot-spots (2000), and indoor localization systems (1999). Under his direction his group has had game changing impact on Microsoft's cloud computing infrastructures both in the datacenter and in wide-area networking. Dr. Bahl is the founder of ACM SIGMOBILE, MobiSys and several other conferences. He is a Fellow of ACM, IEEE, and AAAS. Also, with his wife he co-founded Computing For All, a 501(c)(3) non-profit dedicated to increasing and enhancing computer science education for students of all ages and from all backgrounds.
Abstract: The history of computing has a long track record of working towards persistent, interactive systems that are (ideally) always on and always responsive. Examples include time-sharing systems in the 1970s, embedded systems and databases in the 1980s, web servers in the 1990s, E-commerce systems in the 2000s, and cloud computing and the Internet of Things (IoT) in the 2010s. These systems are distinctly different from personal computers, which spend most of their time shut down or suspended. This property of being "always on" is analogous to living things and requires many of the same mechanisms as living things, including an ability to self repair, an ability to go dormant while still being able to respond quickly to unexpected events, an ability to know one's physical context, and a plastic, learning nervous system. In this talk, I will explore technologies that are evolving towards such capabilities with a particular emphasis on emerging cyber-physical systems, which not only sense their physical context but also manipulate it. I will show that the technology for such "living digital things" is actually co-evolving with a symbiotic human culture that embraces the technology.
Bio: Edward A. Lee is the Robert S. Pepper Distinguished Professor and former chair of the Electrical Engineering and Computer Sciences (EECS) department at U.C. Berkeley. He has been on the faculty at Berkeley since 1986, and has been working on design of embedded systems since 1978. His principal research interest is in the design, modeling, and simulation of distributed, embedded, real-time computational systems. Because of this focus, a major part of his work recently has been on models of computation for dealing with concurrency and timeliness in software. His is well known for his leadership of the Ptolemy Project and the development of software frameworks now known as Ptolemy Classic and Ptolemy II. This project focuses on system-level modeling, simulation, and design, and has codified a taxonomy of approaches to concurrency and communication between software components. One subproject is on models of concurrency in software that are more understandable and analyzable than the prevailing abstractions based on threads and middleware for software components. See http://ptolemy.org for details.
Prof. Lee is also a founding Director of CHESS, the Berkeley Center for Hybrid and Embedded Software Systems. CHESS is a collaboration among several Berkeley faculty focused on building a foundational systems science that is at once computational and physical. The center is developing model-based and tool-supported design methodologies for real-time fault tolerant software on heterogeneous distributed platforms. See http://chess.eecs.berkeley.edu for details.
Satya is the Carnegie Group Professor of Computer Science at Carnegie Mellon University. He received the PhD in Computer Science from Carnegie Mellon, after Bachelor's and Master's degrees from the Indian Institute of Technology, Madras. He is a Fellow of the ACM and the IEEE. He was the founding Program Chair of the HotMobile series of workshops, the founding Editor-in-Chief of IEEE Pervasive Computing, the founding Area Editor for the Synthesis Series on Mobile and Pervasive Computing, and is the founding program chair of the First IEEE Symposium on Edge Computing that will be held in October 2016. He was the founding director of Intel Research Pittsburgh, and was an Advisor to Maginatics , which has created a cloud-based realization of the AFS vision and was acquired by EMC in 2014.
Prof. Lee got his Ph.D. from UC Berkeley in 1986, where his thesis focused on architectures and software for real-time signal processing. His masters is from MIT, where his thesis focused on image processing. His bachelor’s degree is from Yale, where he double majored in Computer Science and Engineering and Applied Science. Between his masters and Ph.D., he spent two years at AT&T Bell Labs, where he worked on embedded software for data communications. He has co-authored five books, including one on embedded and cyber-physical systems, one in digital communications that is in its third edition, one in signals and systems, and one in code generation for dataflow concurrent software.
Prof. Lee is in high demand for invited and plenary talks in a wide variety of forums. He has recently given invited talks and keynote talks at international conferences and workshops (CONCUR, DAC, EC2, EMSOFT, HCSS, HSCC, ICCD, ISORC, ISPCS, LCTES, MEMOCODE, MODELS, ODES, PerMIS, QSIC, RTAS, RTCSA, RTSS, SAMOS, WFCD, WSC etc.), in industry (Bosch, Thales, Microsoft, National Instruments, Xilinx, etc.), and at academic and research distinguished lecture series (Academia Sinica, Arizona State Univ., ARTES Summer School, CMU, Halmstad University, Iowa Sate Univ., Lunds University, National University of Singapore, The Onassis Foundation Science Lecture Series, Purdue, UC Davis, UC Irvine, Virginia Tech, Washington University in St. Louis, etc.).
Prof. Lee is a Fellow of the IEEE and won the Frederick Emmons Terman Award for Engineering Education, 1997 and the NSF Presidential Young Investigator Award in 1987. He was born and raised in San Juan, Puerto Rico, and is fluent in Spanish and capable in French.
Bio: Prof. Amr El Abbadi received his B. Eng. from Alexandria University, Egypt in 1980 and his Ph.D. in Computer Science from Cornell University in 1987. In August 1987 he joined the Department of Computer Science at the University of California, Santa Barbara, where he is currently a Professor. In 1990 he was a visiting professor at the University of Campinas in Brazil, and in 1994 a visiting scientist at IBM Almaden Research Center, in August 1998 a visiting lecturer at the Swedish Institute of Computer Science in Stockholm, Sweden and in July 1999 a visiting researcher at IRISA at the University of of Rennes in France.
He has served as area editor for Information Systems: An International Journal, an editor of Information Processing Letters (IPL) and Associate editor of the Bulletin of the Technical Committee on Data Engineering. He was Vice Chair of the 1999 International Conference on Distributed Computing Systems, Vice Chair for the International Conference on Data Engineering 2002, and the Americas Program Chair for the 2000 International Conference on Very Large Data Bases (VLDB). He received the best-paper award in ICDE 2002, for co-authoring the paper "Exploring Spatial Datasets with Histograms". Prof. El Abbadi's main research interests and accomplishments have been in understanding and developing basic mechanisms for supporting distributed information management systems, including databases, digital libraries, data management of moving objects and geographic information systems. His earlier work concentrated on the development of protocols and algorithms that ensure high availability and fault tolerance in such systems. This body of work mainly focused on the development of concurrency control protocols and replication techniques for managing database systems. More recently, he has extended the domain of research from the traditional database context to more general information management systems that may have weaker or stronger constraints than databases. This involves allocation of data on multiple I/O devices, retrieval of multidimensional data, management and processing of distributed object repositories, and data warehouse as well as developing scalable distributed architectures for locating heterogeneous information sources.