Collaboration with Hong Kong: Minimizing Energy Consumption Through Task Scheduling

这个CNIC奖项支持计划来港旅游。三次TRIPS将用于与香港大学的林德华和香港科技大学的末底改·戈林开展新的研究合作。本文的研究对象是信息技术中普遍存在的能源消耗问题。这三个合作者为这个问题带来了互补的优势。最初的方法是分析任务管理中最小化能耗的策略和算法，使用动态规划将掉电调度和变速调度进行比较，然后探索替代模型的可能性。加州大学河滨分校的一名研究生将参与该项目，并将与PI一起前往香港。预计最初的合作将导致CEISE的全面研究提案。这次计划访问旨在建立一个国际研究合作，可以贡献应对能源管理挑战所需的新见解和算法技术。从理论上讲，它将扩大对掉电和变速调度问题可行解的理解，从而产生更高效的算法。这些调度方法的现有算法目前对于实际应用来说太慢或太复杂。提出的研究旨在设计更快、更简单、更通用的算法。正如一位评价者所指出的那样，在消耗越来越多的能源的信息技术方面采取更有效的办法有助于控制全球能源消耗。因此，这项研究的结果可以对支持可持续计算和可持续的现代社会产生影响。该项目的另一个好处是一名研究生参与了国际合作。

This CNIC award supports planning-visit travel to Hong Kong. Three trips will be used to develop a new research collaboration with Tak-Wah Lam at the University of Hong Kong and Mordecai Golin at the Hong Kong University of Science and Technology. The target of the research is the general problem of energy consumption in information technology. The three collaborators bring complementary strengths to the problem. The initial approach is to analyze strategies and algorithms for minimizing energy consumption in task management, comparing power-down scheduling with speed-scaling scheduling using dynamic programming and then exploring the possibility of alternative models. A graduate student from UC-Riverside will participate in the project and will travel to Hong Kong with the PI. It is anticipated that the initial collaboration will lead to a full-scale research proposal to CISE. This planning visit aims to establish an international research collaboration that can contribute new insights and algorithmic techniques required to address challenges in energy management. From a theoretical point of view it will expand the understanding of feasible solutions of power-down and speed-scaling scheduling problems, leading to more efficient algorithms. Existing algorithms for these scheduling approaches are currently too slow or too complex for practical applications. The proposed research aims to design faster, simpler, and more general algorithms. As one of the reviewers noted, more efficient approaches in information technology, which is consuming increasing amounts of energy, can help to control global energy consumption. Thus, the results of this research can have an impact on supporting sustainable computing and a sustainable modern society. Another benefit of the project is the involvement of a graduate student in the international collaboration.