平均服务器温度约束下的数据中心能耗最小化控制算法

发布时间：2018-10-29 22:33

【摘要】：近几年来,互联网数据中心的能耗问题已成为业界关注的热点。已有的研究工作多数致力于仅降低服务器系统的能耗而忽略机房空调(CRAC)系统的能耗,或没有考虑到服务质量(Qo S)约束和服务器CPU温度约束。研究了满足Qo S和服务器CPU平均温度约束的数据中心能耗最小化问题,提出了基于李雅普诺夫优化理论的能耗最小化算法,以降低数据中心的总能耗。通过构建系统模型,使用数据中心真实工作负载数据对算法性能进行评价,表明该算法可以实现数据中心服务器能耗和CRAC能耗总和的最小化。
[Abstract]:In recent years, the energy consumption of Internet data center has become the focus of attention. Most of the existing researches focus on reducing the energy consumption of the server system but neglecting the energy consumption of the air-conditioning (CRAC) system in the computer room, or not taking into account the quality of service (Qo S) constraint and the server CPU temperature constraint. In this paper, the problem of minimizing the energy consumption of data centers satisfying the average temperature constraints of Qo S and server CPU is studied, and an energy consumption minimization algorithm based on Lyapunov optimization theory is proposed to reduce the total energy consumption of data centers. By constructing the system model and evaluating the performance of the algorithm by using the real workload data of the data center, it is shown that the algorithm can minimize the total energy consumption of the data center server and CRAC.
【作者单位】：内蒙古工业大学信息工程学院;鄂尔多斯云泰互联科技有限公司;
【基金】：国家自然科学基金资助项目(61502255,61540004) 内蒙古自治区高等学校科学基金资助项目(NJZY14064) 内蒙古自治区科学基金资助项目(2014BS0607)
【分类号】：TP13;TP308

【参考文献】