Optimal CPU Frequency Scaling Policies for Sustainable Edge Computing
Optimal CPU Frequency Scaling Policies for Sustainable Edge Computing
复制标题
DOI：
10.1109/pimrc50174.2021.9569613
复制
发表时间：
2021-09
期刊：
影响因子：
--
通讯作者：
Yu Luo;Lina Pu;Chun-Hung Liu
中科院分区：
文献类型：
--
作者：
Yu Luo;Lina Pu;Chun-Hung Liu
可持续边缘计算（SEC）是一项很有前途的技术，可以降低移动物联网（IoT）的能耗和计算延迟。通过从环境中收集太阳能或风能，电网外的SEC云可以为资源受限的移动物联网设备提供强大的计算能力。考虑到可持续能源的密度随时间的显著变化，SEC云let需要动态调整中央处理器（CPU）的时钟频率，以平衡能耗和计算能力。在本文中，我们考虑到云let的能量存储有限，并制定了离线最优CPU频率缩放策略，以最大化云let在一定时间内的整体计算能力。找到了给出最佳CPU频率缩放的图形化观点的最紧字符串策略。
Sustainable edge computing (SEC) is a promising technology that can reduce energy consumption and computing latency for the mobile Internet of things (IoT). By collecting solar or wind energy from the environment, an SEC cloudlet outside the electric grid can provide powerful computing capabilities for resource-constrained mobile IoT devices. Considering significant density variation of sustainable energy over time, the SEC cloudlet needs to dynamically adjust the clock frequency of the central processing unit (CPU) to balance energy consumption and computing power. In this paper, we consider the limited energy storage of the cloudlet and develop an offline optimal CPU frequency scaling policy to maximize the overall computing power of the cloudlet within a certain period of time. The tightest string policy that gives a graphical viewpoint of the optimal CPU frequency scaling is found.