What to do about the end of Moore's law (probably)?
撰稿:
摄影:
发布时间:2012年04月27日
报告时间:2012年5月3日(周四)上午 10:30-11:30
报告地点:计算所 四层报告厅
主讲人:Krishna Palem,NTU, Singapore & Rice University, USA
报告摘要:
Many claim that the laws of physics dictating the exponentially improving benefits of Moore’s Law will end in the next 10 to 20 years. The argument for the end of Moore’s Law is based, in part, on an analysis that switching devices cannot function deterministically as feature sizes get reduced to molecular levels. Moore’s Law could, however, continue provided systems with probabilistic switches could process information usefully.
My research suggests that this is indeed possible in contexts where the "quality" of the results of the computation is perceptually determined by our senses—audio and video information being significant examples. To demonstrate this principle, I will show how CMOS-based devices, circuits and computing architectures whose correctness is characterized probabilistically can be used effectively. I will show that significant (depending on the application, a multiplicative factor of 2 to 30) energy, performance and area gains can be achieved, while trading a perceptually tolerable level of error—through probabilistic adders and multipliers, applied in the context of finite impulse response filters and FFT engines processing video and audio data in digital signal processing. Quantifying the human tolerance for error, we expect, will be ultimately based on neurobiological models.
Conceptually, our thesis recommending tolerating error in the switching devices in return for savings in cost, is analogous to Herb Simon’s notion of satisficing—we will conclude the talk by dwelling on this analog and its implications to probabilistic design of future ultra large-scale integrated (ULSI) circuits and computing architectures.
主讲人简介:
Krishna V. Palem is the Ken and Audrey Kennedy Professor at Rice University with appointments in Computer Science, in Electrical and Computer Engineering and in Statistics. He founded and directs the NTU-Rice Institute on Sustainable and Applied Infodynamics (ISAID), and is a scholar in the Baker Institute for Public Policy. Concurrently, he is a Nanyang Visiting Professor in the School of Physical and Mathematical Sciences at the Nanyang Technological University (NTU), Singapore. He was a Moore Distinguished Faculty Fellow at Caltech in 2006-2007, and a Schonbrunn Fellow at the Hebrew University of Jerusalem in 1999, where he was recognized for excellence in teaching.
What to do about the end of Moore's law (probably)?
报告时间:2012年5月3日(周四)上午 10:30-11:30
报告地点:计算所 四层报告厅
主讲人:Krishna Palem,NTU, Singapore & Rice University, USA
报告摘要:
Many claim that the laws of physics dictating the exponentially improving benefits of Moore’s Law will end in the next 10 to 20 years. The argument for the end of Moore’s Law is based, in part, on an analysis that switching devices cannot function deterministically as feature sizes get reduced to molecular levels. Moore’s Law could, however, continue provided systems with probabilistic switches could process information usefully.
My research suggests that this is indeed possible in contexts where the "quality" of the results of the computation is perceptually determined by our senses—audio and video information being significant examples. To demonstrate this principle, I will show how CMOS-based devices, circuits and computing architectures whose correctness is characterized probabilistically can be used effectively. I will show that significant (depending on the application, a multiplicative factor of 2 to 30) energy, performance and area gains can be achieved, while trading a perceptually tolerable level of error—through probabilistic adders and multipliers, applied in the context of finite impulse response filters and FFT engines processing video and audio data in digital signal processing. Quantifying the human tolerance for error, we expect, will be ultimately based on neurobiological models.
Conceptually, our thesis recommending tolerating error in the switching devices in return for savings in cost, is analogous to Herb Simon’s notion of satisficing—we will conclude the talk by dwelling on this analog and its implications to probabilistic design of future ultra large-scale integrated (ULSI) circuits and computing architectures.
主讲人简介:
Krishna V. Palem is the Ken and Audrey Kennedy Professor at Rice University with appointments in Computer Science, in Electrical and Computer Engineering and in Statistics. He founded and directs the NTU-Rice Institute on Sustainable and Applied Infodynamics (ISAID), and is a scholar in the Baker Institute for Public Policy. Concurrently, he is a Nanyang Visiting Professor in the School of Physical and Mathematical Sciences at the Nanyang Technological University (NTU), Singapore. He was a Moore Distinguished Faculty Fellow at Caltech in 2006-2007, and a Schonbrunn Fellow at the Hebrew University of Jerusalem in 1999, where he was recognized for excellence in teaching.