Tulika Mitra is a Professor of Computer Science at the School of Computing, National University of Singapore (NUS). She received her PhD from the State University of New York at Stony Brook (2000), M.E. from the Indian Institute of Science (1997), and Bachelor’s degree from Jadavpur University (1995)—all in Computer Science. Her research interests span various aspects of the design automation of embedded real-time systems, cyber-physical systems (CPS), and the Internet of Things (IoT), with particular emphasis on energy-efficient computing, heterogeneous computing, application-specific processors, and software timing analysis/optimizations. She has authored more than one hundred scientific publications in leading international journals and conferences in this domain.
Prof. Mitra currently serves as the Deputy Editor-in-Chief of IEEE Embedded Systems Letters, Senior Associate Editor of the ACM Transactions on Embedded Computing Systems, Associate Editor of IEEE Design & Test Magazine, EURASIP Journal on Embedded Systems. She has served as Associate Editor of IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems and IET Computers & Digital Techniques in the past and on the organizing and program committees of several major conferences in embedded systems, real-time systems, and electronic design automation.
National University of Singapore, Singapore
Phone: +65-365166839
URL: http://www.comp.nus.edu.sg/~tulika/
tulika@comp.nus.edu.sg
DVP term expires December 2020
Presentations
Behind the Scenes of the Internet-of-Things Revolution
The Internet of Things (IoT), where physical objects or “things” embedded with computing power and sensors connect to the network for seamless cooperation between the cyber domain and the physical world, is poised to improve all aspects of our lives dramatically. By 2020, an estimated 50 billion IoT devices will empower mankind to collect, mine, and analyze an astronomical amount of data towards information, knowledge, and real-time responses in various sectors from healthcare, transportation to agriculture, energy, manufacturing. The confluence of trends driving the IoT revolution from the computing perspective will be presented, current impediments thwarting its progress will be discussed, and the exciting technological advances that are expected to overcome these obstacles in realizing its full potential will be introduced.
Mobile Heterogeneous Computing: A Software Perspective
Mobile heterogeneous computing, in the form of the multiprocessor system-on-chips (MPSoC)—composed of various processing elements such as general-purpose cores with differing characteristics (GPUs, DSPs, non-programmable accelerators, and reconfigurable computing)—are expected to dominate the current and the future mobile platform landscape. The heterogeneity enables a computational kernel with specific requirements to be paired with the processing element(s) ideally suited to perform that computation, leading to substantially improved performance and energy-efficiency. While heterogeneous computing is an attractive proposition in theory, considerable software support at all levels is essential to fully realize its promises.
The system software needs to orchestrate the different on-chip compute resources in a synergistic manner with minimal engagement from the application developers. The current state-of-the-art is inadequate in the software dimension despite tremendous progress and success in designing heterogeneous MPSoCs for mobile devices.
This talk will put the spotlight on the software perspective of mobile heterogeneous computing, especially in the context of popular emerging applications, such as 3D gaming, multimedia processing and analytics. The talk will introduce the technology trends driving the mobile heterogeneous computing revolution, provide an overview of computationally and performance divergent compute elements, and present efforts at compiler and runtime management layers to unleash its potential towards high-performance energy-efficient computing.
Accelerators for Smart IoT Devices
At present, the IoT devices in the edge are primarily responsible only for collecting and communicating the data to the cloud, whereas the computationally intensive data analytics takes place in the back-end cloud server. However, the data privacy and the connectivity issues—in conjunction with the fast real-time response requirement of certain IoT applications—call for smart edge devices. These devices should be able to support privacy-preserving, time-sensitive computation for machine intelligence on-site and rely on the cloud only for offline data processing away from the critical path. The prevalent general-purpose processors in the edge devices are not equipped with sufficient compute power or energy-efficiency to provide the necessary computational intelligence.
This talk will present the design of ultra-low power accelerators that overcome these challenges and speed up computation to realize the edge analytics vision.
