12:00 AM - 11:15 AM on Friday, November 1, 2013
Location: Torg 2150
by Dr. Ayse Coskun, Boston University
image Energy efficiency is a central issue in all computing domains. In data centers, operational and cooling costs impose significant sustainability challenges. In tandem, future processors are expected to run complex, highly performance demanding workloads, making the well-studied energy management policies inadequate. High power densities also increase the on-chip temperatures and thermal variations, both of which degrade system reliability and add to the system design complexity. Achieving orders of magnitude of energy efficiency improvements requires novel system and software design approaches coupled with dynamic techniques that recognize the hardware-software characteristics and understand the complex interplay among performance, energy, and temperature. This talk will discuss two closely-tied research thrusts: (1) designing novel 3D stacked architectures and the necessary runtime management strategies for improving processor energy efficiency; and (2) developing workload management and power regulation methods in data centers to reduce the overall energy cost of computing.
Ayse K. Coskun is an assistant professor in the Electrical and Computer Engineering Department at Boston University. She received her MS and PhD degrees in Computer Science and Engineering from University of California, San Diego. Coskun’s research interests are temperature and energy management, 3D stack architectures, computer architecture, embedded systems, and data center energy efficiency. Prof. Coskun worked at Sun Microsystems (now Oracle), San Diego prior to her current position at BU. She received the best paper award at IFIP/IEEE VLSI-SoC Conference in 2009 and at High Performance Embedded Computing (HPEC) Workshop in 2011, and she is a recipient of the NSF CAREER award. She has served as an associate editor for ACM Transactions on Design Automation of Electronic Systems and IEEE Embedded Systems Letters. Coskun also writes a bi-monthly column on green computing for the Circuit Cellar magazine.