A Mobility-Aware Optimal Resource Allocation Architecture for Big Data Task Execution on Mobile Cloud in Smart Cities

Abstract

In recent years, the smart city concept, which involves multiple disciplines, for example, smart healthcare, smart transportation, and smart community, has become popular because of its ability to improve urban citizens' quality of life. However, most services in these areas of smart cities have become data-driven, thus generating big data that require seamless real-time access, sharing, storing, processing, and analysis anywhere at any time for intelligent decision making to improve living standards. In this scenario, MCC can play a vital role by allowing a mobile device to access and offload big-data-related tasks to powerful cloudlet servers attached to many wireless APs, thus ensuring that the QoS demands of end users are met. However, the connectivity of mobile devices with a given AP is not continuous, but rather sporadic with varying signal strengths. Furthermore, the heterogeneity of the cloudlet resources and the big data application requests place additional challenges in making optimal code execution decision. To cope with this problem, this article proposes a mobility- aware optimal resource allocation architecture, namely Mobi-Het, for remote big data task execution in MCC that offers higher efficiency in timeliness and reliability. The system architecture and key components of the proposed resource allocation service are presented and evaluated. The results of experiments and simulations have demonstrated the effectiveness and efficiency of the proposed Mobi-Het architecture for mobile big data applications.