Thermal Estimation and Prediction Techniques for Dynamic Thermal Management of High-Performance Processors
撰稿：
摄影：
发布时间：2013年08月13日
时间：2013年8月15日（周四）上午 10:30-11:30
地点：446会议室
摘要
Temperature estimation and prediction are crucial for on-line regulation of temperature variations and hot spots on today’s high performance processors. In this talk, a new method is presented to accurately estimate and predict the full-chip temperature at runtime under more practical conditions where there are inaccurate power estimations and limited number of on-chip physical thermal sensors. A thermal sensor placement method is also proposed to enhance the performance of thermal estimation and prediction. Experimental results on standard SPEC benchmarks demonstrate the efficiency and accuracy of the new method compared with traditional kalman filter based methods.
主讲人简介
Dr. Hai Wang received his B.S. degree from Huazhong University of Science and Technology in 2007, and his M.S. and Ph.D. degree from University of California, Riverside in 2008 and 2012, respectively. After that, he has been an associate professor at University of Electronic Science and Technology of China. His research interests mainly lie in verification and optimization of VLSI circuits and systems. He serves as TPC member of several international conferences including ASP-DAC and ISQED, and also serves as reviewer of many journals including IEEE TC, IEEE TCAD, and ACM TODEAS.
附件下载：
Thermal Estimation and Prediction Techniques for Dynamic Thermal Management of High-Performance Processors
地点：446会议室
摘要
Temperature estimation and prediction are crucial for on-line regulation of temperature variations and hot spots on today’s high performance processors. In this talk, a new method is presented to accurately estimate and predict the full-chip temperature at runtime under more practical conditions where there are inaccurate power estimations and limited number of on-chip physical thermal sensors. A thermal sensor placement method is also proposed to enhance the performance of thermal estimation and prediction. Experimental results on standard SPEC benchmarks demonstrate the efficiency and accuracy of the new method compared with traditional kalman filter based methods.
主讲人简介
Dr. Hai Wang received his B.S. degree from Huazhong University of Science and Technology in 2007, and his M.S. and Ph.D. degree from University of California, Riverside in 2008 and 2012, respectively. After that, he has been an associate professor at University of Electronic Science and Technology of China. His research interests mainly lie in verification and optimization of VLSI circuits and systems. He serves as TPC member of several international conferences including ASP-DAC and ISQED, and also serves as reviewer of many journals including IEEE TC, IEEE TCAD, and ACM TODEAS.