Thermodynamics and Computation: Towards a New Synthesis

热力学和计算：迈向新的合成

• 批准号: 1740919
• 负责人: David Wolpert
• 金额: $ 4.5万
• 依托单位: Santa Fe Institute
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-07-01 至 2018-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1740919&HistoricalAwards=false
• 关键词: Thermodynamics; Computation; Towards; New; Synthesis

This workshop will bring researchers from various science fields together at the Santa Fe Institute for three days to foster development of a new synthesis of thermodynamics and computation. Researchers from non-equilibrium statistical mechanics, theoretical computer science, and those in computer engineering focused on energy-efficient computing will participate in meeting. Approximately 5% of US energy consumption is used to run computers and this energy bill accounts for a large fraction of the lifetime budget of a modern high-performance computing center. Improving the energy efficiency of computers is crucial for reducing the total energy usage, which is a focus of this workshop. Using recent developments in non-equilibrium statistical mechanics, now is the time for a new synthesis between physics, theory of computing, and computer engineering to make great strides in energy-efficient computing. Graduate students and postdocs will participate, building up the next generation of researchers in this highly cross-disciplinary, nascent area of research. Since Landauer and Bennett's pioneering analysis, none of the theoretical analyses of energy-efficient computing has yet taken advantage of the new understanding coming from non-equilibrium statistical mechanics, such as partially reversible computing, approximate computing, and average-case complexity. Computational scientists and computer engineers that have focused their effort on phenomenological, practical successes in energy-efficient computing will participate in the workshop. Insights from the fundamental physics will suggest new opportunity areas for the practice of energy-efficient computing. Computer scientists participating in the meeting will interact with the rest of the scientists to understand which are the most relevant thermodynamical questions to determine the next directions in energy-efficient and large-scale computing.

该研讨会将把来自不同科学领域的研究人员聚集在圣达菲研究所为期三天，以促进热力学和计算的新综合发展。来自非平衡统计力学、理论计算机科学以及专注于节能计算的计算机工程领域的研究人员将参加会议。美国大约 5% 的能源消耗用于运行计算机，而这笔能源费用占现代高性能计算中心生命周期预算的很大一部分。提高计算机的能源效率对于减少总能源使用量至关重要，这也是本次研讨会的重点。利用非平衡统计力学的最新发展，现在是物理学、计算理论和计算机工程之间新的综合，以在节能计算方面取得巨大进步的时候了。研究生和博士后将参与其中，在这个高度跨学科的新兴研究领域培养下一代研究人员。自从兰道尔和贝内特的开创性分析以来，节能计算的理论分析还没有利用来自非平衡统计力学的新理解，例如部分可逆计算、近似计算和平均情况复杂性。致力于在节能计算方面取得现象学和实际成功的计算科学家和计算机工程师将参加研讨会。来自基础物理学的见解将为节能计算的实践提供新的机会领域。参加会议的计算机科学家将与其他科学家互动，了解哪些是最相关的热力学问题，以确定节能和大规模计算的下一步方向。