CNS Core: Small: Managing Electrical and Thermal Energy in Sustainable Computing Systems

CNS 核心：小型：管理可持续计算系统中的电能和热能

• 批准号: 2230143
• 负责人: David Irwin
• 金额: $ 32.6万
• 依托单位: University of Massachusetts Amherst
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2230143&HistoricalAwards=false
• 关键词: CNS; Core; Small; Managing; Electrical

Sustainable computing systems operate on zero-carbon renewable energy harvested from their environment, such as solar or wind, and stored in batteries. Importantly, renewable-powered systems may be deployed in many different climates that subject them to a wide range of ambient temperatures that significantly alter the efficiency and correctness of these systems' underlying components, including their processors, batteries, solar cells, and clocks. Unfortunately, current systems are either designed for a narrow and ideal temperature range, and thus are often unreliable under even slight temperature variations, or must consume significant additional energy to maintain an ideal temperature within a narrow window, which significantly reduces their energy-efficiency. To address the problem, this project proposes fundamental research on the design of sustainable renewable-powered systems that are ectothermic in that they jointly manage and adapt to variations in the electrical and thermal energy available in their environment to optimize their energy-efficiency, performance, and reliability. While the temperature responses of individual system components are well-known, ectothermic design takes a holistic systems approach that exploits relationships between components and their environment to co-optimize system-wide energy-efficiency, correctness, performance, and reliability. To this end, this project will develop methods for understanding, modeling, and exploiting the dependencies between computation, heat generation, the ambient environment, renewable energy availability, and the workload. This project includes numerous broader impacts. The project has the potential for significant societal impact and positively address the climate concerns of computing by improving the energy-efficiency, correctness, performance, and reliability of sustainable renewable-powered computer systems. Our project also has the potential for technical impact by improving the design of renewable-powered systems at all scales--from small embedded platforms to large data centers. This project plans to engage in multiple educational activities, including giving tutorials that focus on the relationship between computing and energy consumption at UMass summer programs for high school students in engineering and computing, and integrating ectothermic design into graduate and undergraduate courses. Finally, the project will emphasize the recruitment of students from under-represented groups in computing.This proposal is funded in part from the DCL on Design for Sustainability in Computing (NSF-22-060)This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

可持续计算系统依靠从其环境中收集的零碳可再生能源（如太阳能或风能）运行，并储存在电池中。 重要的是，可再生能源发电系统可以部署在许多不同的气候条件下，这些气候条件会使它们受到各种环境温度的影响，这些环境温度会显著改变这些系统底层组件的效率和正确性，包括它们的处理器，电池，太阳能电池和时钟。不幸的是，当前的系统要么是针对窄且理想的温度范围设计的，因此即使在轻微的温度变化下也常常是不可靠的，要么必须消耗显著的额外能量以在窄窗口内维持理想温度，这显著降低了它们的能量效率。 为了解决这个问题，该项目提出了对可持续可再生能源系统设计的基础研究，这些系统是吸热的，因为它们共同管理和适应环境中可用的电能和热能的变化，以优化其能源效率，性能和可靠性。虽然单个系统组件的温度响应是众所周知的，但吸热设计采用整体系统方法，利用组件及其环境之间的关系来共同优化系统范围的能效，正确性，性能和可靠性。 为此，该项目将开发用于理解，建模和利用计算，发热，周围环境，可再生能源可用性和工作负载之间的依赖关系的方法。 该项目包括许多更广泛的影响。 该项目有可能产生重大的社会影响，并通过提高可持续可再生能源计算机系统的能效、正确性、性能和可靠性，积极解决计算的气候问题。我们的项目还具有技术影响的潜力，可以改善各种规模的可再生能源系统的设计-从小型嵌入式平台到大型数据中心。 该项目计划参与多种教育活动，包括在麻省大学工程和计算专业的高中生暑期课程中提供专注于计算和能源消耗之间关系的教程，并将光热设计融入研究生和本科生课程。最后，该项目将强调从计算领域代表性不足的群体中招收学生。该提案部分由DCL on Design for Sustainability in Computing（NSF-22-060）资助。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Jointly Managing Electrical and Thermal Energy in Solar- and Battery-powered Computer Systems

联合管理太阳能和电池供电计算机系统中的电能和热能

• DOI: 10.1145/3575813.3595191
• 发表时间: 2023
• 期刊: 14th ACM International Conference on Future Energy Systems
• 作者: Bashir, Noman;Chandio, Yasra;Irwin, David;Anwar, Fatima M.;Gummeson, Jeremy;Shenoy, Prashant
• 通讯作者: Shenoy, Prashant

HoloSet - A Dataset for Visual-Inertial Pose Estimation in Extended Reality: Dataset

HoloSet - 扩展现实中视觉惯性姿态估计的数据集：数据集

• DOI: 10.1145/3560905.3567763
• 发表时间: 2022
• 期刊: ACM Conference on Embedded Networked Sensor Systems
• 作者: Chandio, Yasra;Bashir, Noman;Anwar, Fatima M.
• 通讯作者: Anwar, Fatima M.