CSR: Small: Improving Data Center Water Efficiency via Online Resource Management

大型数据中心可能每天消耗数百万加仑的冷却水；此外，数据中心还间接消耗异地发电中嵌入的大量水。因此，在预计全球用水需求激增的情况下，节水正成为数据中心的一个关键问题。如果不加以控制，数据中心日益增长的水足迹可能会对数据中心的可持续性构成严重威胁，甚至可能阻碍服务的可用性，特别是对于缺水地区的数据中心。现有的机械保护解决方案，如使用回收水/工业用水和直接使用室外冷空气，往往昂贵和/或非常受限于外部因素，如地点、气候条件等。作为业界和学术界为实现数据中心可持续性而做出的整体努力的一部分，该项目独特地将水足迹作为虚拟数据中心资源管理的重要组成部分进行了整合。它利用了水效随时间变化的固有但鲜为人知的特点，并优化了资源管理，在不影响服务质量的情况下将运营成本和水足迹降至最低。为此，本项目研究了三个相辅相成的研究方向：(I)在存在大量未知动态(如外部温度高度挥发)的情况下，针对水可持续性的在线计算资源管理；(Ii)探索数据中心水效对冷却系统的依赖关系，以及针对水可持续性的冷却和计算资源管理的联合优化；(Iii)基于系统原型和软件模拟相结合的实验和验证。除了对环境的影响外，该项目还具有巨大的社会影响，因为它有可能有效和经济地缓解数据中心水占用的日益增长的压力。该项目将提高对数据中心用水量、电力/能源消耗和资源管理之间相互影响的理解，为更绿色的数据中心的可持续发展奠定坚实的基础。新的发现和技术将很容易被纳入教材。该项目还包含一个重要组成部分，通过激励少数族裔学生，特别是拉美裔学生从事计算机科学研究来促进多样性。

A large data center may consume millions of gallons of cooling water each day; in addition, data centers also indirectly consume an enormous amount of water embedded in offsite electricity generation. As a result, water conservation is surfacing as a critical concern for data centers, amid the anticipation of surging water demand worldwide. Left unchecked, the growing water footprint of data centers can pose a severe threat to data center sustainability and may even handicap availability of services, especially for data centers in water-stressed areas. Existing mechanical solutions for conservation, such as using recycled/industry water and directly using outside cold air, are often costly and/or very limited by external factors such as locations, climate conditions, among others. As part of the integral efforts from both industry and academy to enable data center sustainability, this project uniquely integrates water footprint as an essential part of resource management in virtualized data centers. It exploits the inherent yet little-known characteristics of time-varying water efficiency and optimizes resource management for minimizing operational cost as well as water footprint without compromising service quality. To this end, this project investigates three complementary research thrusts: (i) Online computing resource management for water sustainability in the presence of large unknown dynamics (such as highly volatile outside temperature); (ii) Exploration of the dependency of data center water efficiency on cooling systems, and joint optimization of cooling and computing resource management for water sustainability; (iii) Experimentation and validation based on combined system prototyping and software simulation. In addition to its environmental impacts, this project has large societal impacts for its potential to alleviate the growing pressure on data center water footprint effectively and economically. This project will enhance the understanding of interplay among water consumption, power/energy consumption, and resource management in data centers, and lay a solid foundation for sustainable evolvement of greener data centers. New findings and techniques will be readily incorporated into teaching materials. The project also contains a significant component to promote diversity by inspiring minority students, especially Hispanic students, to engage in computer science research.