学术报告:Energy-Efficient Computing through Integration with Emerging T...
来源: 发布日期:2015-06-08
学 术 报 告
题 目:Energy-Efficient Computing through Integration with Emerging Technologies
主 讲 人:傅昕 博士 美国休斯顿大学电子与计算机工程系
时 间:2015年6月10日(周三),上午10:30
地 点:首师大北二区二层小会议室
主 办 单 位:首都师范大学信息工程学院
主讲人介绍:
Dr. Xin Fu recently joined the Electrical and Computer Engineering Department, the University of Houston, Houston as an Assistant Professor. From 2010 to 2014 she was an Assistant Professor at the Department of Electrical Engineering and Computer Science, the University of Kansas, Lawrence, and from 2009 to 2010 she was an NSF Computing Innovation Fellow with the Computer Science Department, the University of Illinois at Urbana-Champaign, Urbana. She received her Ph.D. degree in Computer Engineering from the University of Florida, Gainesville, in 2009. Her research interests include computer architecture, energy-efficient computing, high-performance computing, hardware reliability and variability, and mobile computing. Dr. Fu is a recipient of 2014 NSF Faculty Early CAREER Award, 2012 Kansas NSF EPSCoR First Award, and 2009 NSF Computing Innovation Fellow.
内 容 介 绍:
The number of transistors per unit of die area continues to increase, but the historical decrease of transistor supply voltage has largely stalled, leading to an ever-increasing power density. The maximum processor power consumption must always remain within a reasonable envelope due to physical constraints. Therefore, modern processors must become inherently more energy efficient to avoid hitting the “power wall.” Recently, to meet the increasing demands on performance and energy-efficiency, emerging technologies (e.g., non-volatile memory, Tunneling Field Effect Transistors (TFET)) are being deployed in the hardware design. This provides chip designers new tools to address the serious power challenge in future processors.
In this talk, Xin will discuss energy-efficient computing in various architecture designs by integrating these emerging technologies. She will present the hybrid TFET-CMOS-based register file design to maximize the power saving without performance loss in throughput processors. Next she will discuss the introduction of non-volatile memory into the embedded processors. Finally, she will present the investigation on heterogeneous many-core CPUs.