Energy-Aware Distributed Multi-Cloud Flower Pollination Optimization Scheme

Abstract

Cloud computing has rapidly emerged as a successful paradigm for providing ICT infrastructure, resources, and services on a pay-per-use basis over the past few years. The wider adoption of Cloud and virtualization technologies has led to the establishment of large-scale datacenters that consume excessive energy. This paper proposed Energy-Aware distributed MultiCloud Flower Pollination Optimization (EAdM-FPO) scheme for minimizing energy consumption and SLA violation with higher resource utilization in a distributed datacenters that belong to a single service provider located in different environments using centralized architecture. The proposed scheme overcome the limitation of using single Cloud datacenter. Four mechanisms namely; Dynamic Switching Probability (DSP), Power Usage Effectiveness (PUE), clustering, and migration controller have been incorporated into the EAdM-FPO scheme to further improve resource management and to reduce the computational complexity of the original FPA. EAdM-FPO scheme has been evaluated based on average energy consumption, VM migration, and SLA violation. In addition, the performance of the proposed EAdM-FPO scheme has been evaluated and compared with related existing schemes based on the same metrics and configuration. Thus, the EAdM-FPO scheme significantly reduce the energy consumption and increase the datacenter resource utilization with less Service Level Agreement (SLA) violation.