Performance evaluation of dynamic cloud resource migration based on temporal and capacity-aware policy for efficient resource sharing

Abstract

This paper elaborates on practical considerations, such as location and capacity issues to offload resources, by adopting a rack based approach for the implementation. The proposed cooperative migration of resources enables efficient resource manipulation without any intermittent execution of the claimed tasks by the mobile devices, while it significantly reduces crash failures that lead all servers to become unavailable within a rack. In addition, this paper presents a modular resource migration scheme for failure-aware resource allocation, where according to the estimated performance of the resource sharing process (e.g. access time, service delay etc.) resources are migrated to another cloud rack based on the associated performance-oriented metrics. The proposed architecture is thoroughly evaluated through simulation tests for the resource migration policy used in the context of cloud rack failures for delay-bounded resource availability of mobile users, as well as for the efficiency of the proposed resource migration scheme.