Efficiency-Scalable Servers for Large-Scale Distributed Sensing Data
Event details
| Date | 12.05.2014 |
| Hour | 14:00 › 15:30 |
| Speaker | Dr. Phillip Stanley-Marbell |
| Location | |
| Category | Conferences - Seminars |
Distributed sensing systems generate lots of data. Instead of moving this data to centralized servers for processing, we can instead design compute servers for processing the data “in the field”. Doing so however requires balancing the conflicting demands of frugal energy usage, high-performance, and, oftentimes, permissible inaccuracy of computation. In this talk, I will address these challenges by looking at them from the ground up, focusing on two contributions. First, I will introduce new results on the energy-efficiency of parallelism as a function of architectural and semiconductor-device properties, and metrics that should be used in characterizing system behavior under a variety of invariance constraints. Next, these analyses will be extended to the modeling of large distributed compute and sensing systems. The insight gained from the analyses will be used to motivate two generations of embedded multiprocessor research platforms for processing large-scale distributed sensing data.
Bio: Phillip Stanley-Marbell received his Ph.D. from Carnegie Mellon University in 2007. He was a post-doctoral researcher at TU Eindhoven until 2008, when he joined IBM Research-Zurich as a permanent Research Staff Member. In 2012 he joined Apple to see his ideas deployed in real-world products. Prior to completing his Ph.D., he held intern and full-time positions at Bell-Labs, Philips, Lucent's Data Networking Group, and NEC Research Labs.
Dr. Stanley-Marbell is the author of a programming language textbook published by John Wiley & Sons in 2003, and of over thirty scientific publications and seven patents / patent applications. He is the developer of the Sunflower embedded multiprocessor instruction-level simulator and the Sunflower sensor hardware platforms, both open source and available online at http://sflr.org. Dr. Stanley-Marbell is a member of the ACM, IEEE, Sigma Xi, USENIX, and the Swiss Mathematical Society. From 2003--2004, he served as the copy editor for the ACM Mobile Computing and Communications Review journal.
Bio: Phillip Stanley-Marbell received his Ph.D. from Carnegie Mellon University in 2007. He was a post-doctoral researcher at TU Eindhoven until 2008, when he joined IBM Research-Zurich as a permanent Research Staff Member. In 2012 he joined Apple to see his ideas deployed in real-world products. Prior to completing his Ph.D., he held intern and full-time positions at Bell-Labs, Philips, Lucent's Data Networking Group, and NEC Research Labs.
Dr. Stanley-Marbell is the author of a programming language textbook published by John Wiley & Sons in 2003, and of over thirty scientific publications and seven patents / patent applications. He is the developer of the Sunflower embedded multiprocessor instruction-level simulator and the Sunflower sensor hardware platforms, both open source and available online at http://sflr.org. Dr. Stanley-Marbell is a member of the ACM, IEEE, Sigma Xi, USENIX, and the Swiss Mathematical Society. From 2003--2004, he served as the copy editor for the ACM Mobile Computing and Communications Review journal.
Practical information
- Informed public
- Free
Organizer
- Babak Falsafi
Contact
- Stéphanie Baillargues