Resource Management in Sustainable Cyber-Physical Systems Using Heterogeneous Cloud Computing

Abstract

The substantial growth of the distributed computing using heterogeneous computing has enabled great expansions in <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Cyber Physical Systems</i> (CPS). Combining CPS with heterogeneous cloud computing is an alternative approach for increasing sustainability of the system. However, execution of resource management in cloud systems is still encountering a few challenges, including the bottlenecks of the Web server capacities and task assignments in the heterogeneous cloud. The unstable service demands often result in service delays, which embarrasses the competitiveness of the enterprises. This paper addresses the problem of the task assignment in heterogeneous clouds, which is proved as a NP-hard problem. The proposed approach is called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Smart Cloud-based Optimizing Workload (SCOW) Model</i> that uses predictive cloud capacities and considers sustainable factors to assign tasks to heterogeneous clouds. To reach the optimization objective, we propose a few algorithms, which include <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Workload Resource Minimization Algorithm</i> (WRM), <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Smart Task Assignment (STA) Algorithm</i> , and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> Task Mapping Algorithm</i> (TMA). Our experimental evaluations have examined the performance of the proposed scheme.