Energy Efficient Optimization with Threshold Based Workflow Scheduling and Virtual Machine Consolidation in Cloud Environment

Cloud computing provides users with usage-based IT services on-demand basis. In these cloud centers, physical machines (PMs) are combined with virtual machines (VMs). Improper planning in workflow scheduling and VM consolidation disturbs the load balancing capability of the system thereby reducing the overall energy of the system with rapid increase in execution time. In this paper, the energy-efficient multi-objective adaptive Manta ray foraging optimization (MAMFO) is proposed for efficient workflow planning. It also optimizes the multi-objective factors such as energy consumption and resource utilization, i.e., CPU and memory. Dynamic Threshold with Enhanced Search and Rescue (DT-ESAR) is introduced for the VM Consolidation System. The dynamic threshold identifies the hosts that are underutilized, overutilized, and normalized. ESAR migrates the VMs from one host to another based on the threshold number. The proposed framework improves energy efficiency and minimizes the time span of the process flow. The experimental results show the efficiency of the proposed approach in terms of energy consumption, makespan, number of migrations and overall SLA. The proposed framework energy consumption is 0.234 kWh, the makespan is 107.25, the number of VM migrations performed is 51, and the overall SLA is 5.23. To determine whether the proposed MAMFO/DT-ESAR method is effective, the findings are compared with the existing methods. Utilizing CloudSim for the experimental evaluation, it is found that the suggested approach significantly improved resource utilization and energy efficiency.