学术

NVIDIA(官方中文名称英伟达)公司Yakun SHAO研究员学术报告

编辑:徐丽君 来源:信息与电子工程学院 时间:2017年09月12日 访问次数:175  源地址

题目: Towards Architectures for the Next  Decade: Modeling and Design of Hardware Accelerators
时间: 20170914日上午 11:00-12:00
地点: 玉泉校区信电楼215会议室
报告人: Yakun SHAO 研究员
Abstract:
     Traditional performance and  energy scaling benefits based on technology improvements have slowed greatly. At  the same time, the demand for computing capability is unsatisfied with new  killer applications emerging in the domains of robotics, automotive, and machine  intelligence. Lack of progress in technology scaling will necessarily place more  demands on the computer architecture and software layers to deliver higher  capability.
     To overcome these challenges,  hardware acceleration in the form of data path and control circuitry customized  to particular algorithms or applications has surfaced as a promising approach,  as it delivers orders of magnitude performance and energy benefits compared to  general-purpose solutions. The importance of accelerators is already evident in  today's energy-efficient mobile SoCs and network processors. These  accelerator-centric architectures require quite different tools and design  methodologies from general-purpose designs. In this talk, I will discuss some of  my PhD work on developing new methodologies for workload characterization, rapid  accelerator design, pre-RTL simulation, and system integration of  accelerator-centric systems.
Biography:
Yakun SHAO is a Research  Scientist at NVIDIA. She received her Ph.D. degree in Computer Science from  Harvard University in 2016, working with Professor David Brooks and Professor  Gu-Yeon Wei. Her primary research interests lie in the general area of computer  architecture, with a particular focus on modeling and design for heterogeneous  architectures. She received her B.S. degree in Electrical Engineering from  Zhejiang University, China and her S.M. degree in Computer Science from Harvard  University. Her previous experience includes internships at Intel Labs and IBM  T.J Watson Research Center. Her paper was selected as one of the Top Picks in  Computer Architecture in 2015. She is a Siebel Scholar and a recipient of the  IBM Ph.D. Fellowship.


总访问量:10743231