CS Colloquium Series @ UCY
Department of Computer Science - University of Cyprus
The Department of Computer Science at the University of Cyprus holds research colloquiums and social hours approximately once weekly. All university students, faculty, and staff are invited to attend. Notifications about new and upcoming events are automatically disseminated to a variety of institutional lists.
If you don't receive these notifications, but want to get informed about upcoming colloquium announcements, you can do the following:
List
RSS
Directions
Colloquium Coordinator: Demetris Zeinalipour
Colloquium: Building a Trillion Sensors IoT: From Sensors to Servers, Dr. Shidhartha Das (Arm Research Cambridge, UK), Thursday, November 7, 2019, 10:00-11:00 EET.
The Department of Computer Science at the University of Cyprus cordially invites you to the Colloquium entitled:
Building a Trillion Sensors IoT: From Sensors to Servers
 |
Speaker: Dr. Shidhartha Das |
Abstract:
The Internet of Things (IoT) powered by a multitude of interconnected computing devices will make the future of computing pervasive and ubiquitous. Sensors lie at the heart of the IoT acting as the interface between the physical world that we interact and the digital world that we compute with. Applications enabled by the IoT rely upon provisioning of services and analytics through a hierarchy of sensor-nodes that then relay useful information to the cloud for further processing.
In this talk, we argue that the design of an energy-efficient IoT is necessarily underpinned by efficiency gains accrued across the entire computing stack – from sensors to server-systems. At the high-end, we make the argument that effectively managing design margins is crucial to delivering energy-efficient computation in an era where efficiency gains through Moore’s Law scaling has effectively stalled. We focus on on-chip sensors and how by sensing their own operating environments, chips can automatically adapt and tune themselves to efficient operating points. We also briefly discuss a novel approach to characterizing computing platforms, based on radiated EM spectra, developed in collaboration with University of Cyprus.
At the low-end of the computing spectrum, we make the case that truly ubiquitous computation can be achieved by embedding greater computational capabilities in power-constrained sensor-nodes. To this end, we examine how emerging devices (particularly, non-volatile memory technologies) can be harnessed to accelerate machine-learning algorithms with orders-of-magnitude improvements in energy-efficiency. Finally, we conclude the talk with an observation that future improvements in computational efficiency will stem from a co-design approach where process-technology innovations are effectively combined with innovative circuits and system design.
Short Bio:
Shidhartha Das is a Senior Principal Research Engineer in Arm Research, based in Cambridge, UK. Dr. Das received his M.Sc. and Ph.D. from University of Michigan in 2003 and 2009, respectively. His research interests include emerging non-volatile memory technologies, micro-architectural and circuit design for variation measurement and mitigation, on-chip power delivery and VLSI architectures for digital signal processing (DSP) accelerators.
Dr. Das holds 46 US granted patents and is the inventor on several more that are pending. He is the recipient of the Arm “Patent Cube” in 2017 and the Arm “Inventor of the Year” award in 2016 for his contributions to emerging non-volatile memory technologies. Dr. Das is the recipient of multiple best-paper awards (CAL 2017, ISLPED 2015, SAME 2010, MICRO 2003), best-paper nominations (ISLPED 2015) and the Microprocessor Review Analysts’ Choice Award in Innovation in 2007. Dr. Das has served as a Guest-Editor for the Journal of Solid-State Circuits (JSSC) and Associate-Editor for IEEE Solid-State Circuits Letters (SSCL). He serves on the Technical Program Committee of ESSCIRC and the ISSCC Students Research Preview Committee.
Note:
The colloquium is jointly organized by the Department of Computer Science at the University of Cyprus and the Department of Electrical and Computer Engineering at the University of Cyprus.
