An Energy-Efficient Resource Management System for a Mobile Ad Hoc Cloud

IEEE Access

Sayed Chhattan Shah¹ 

[1] Department of Information Communication Engineering, Hankuk University of Foreign Studies, Seoul, South Korea;

关键词: Cloud robotics;    local data center;    mobile cloud;    resource scheduling;    sensor cloud;   

DOI  :  10.1109/ACCESS.2018.2876600

来源: DOAJ

【 摘 要 】

Recently, mobile ad hoc clouds have emerged as a promising technology for mobile cyber physical system applications, such as mobile intelligent video surveillance and smart homes. Resource management plays a key role in maximizing resource utilization and application performance in mobile ad hoc clouds. Unlike resource management in traditional distributed computing systems, such as clouds, resource management in a mobile ad hoc cloud poses numerous challenges owing to the node mobility, limited battery power, high latency, and dynamic network environment. The real-time requirements associated with mobile cyber physical system applications make the problem even more challenging. Currently existing resource management systems for mobile ad hoc clouds are not designed to support mobile cyber physical system applications and energy-efficient communication between application tasks. In this paper, we propose a new energy-efficient resource management system for mobile ad hoc clouds. The proposed system consists of two layers: a network layer and a middleware layer. The network layer provides ad hoc network and communication services to the middleware layer and shares the collected information in order to allow efficient and robust resource management decisions. It uses: (1) a transmission power control mechanism to improve energy efficiency and network capacity; (2) link lifetimes to reduce communication and energy consumption costs; and (3) link quality to estimate data transfer times. The middleware layer is responsible for discovery, monitoring, migration, and the allocation of resources. It receives application tasks from users and allocates tasks to nodes on the basis of networkand node-level information.

【 授权许可】

Unknown