EELB: an energy-efficient load balancing model for cloud environment using Markov decision process

Cloud computing is termed as an on-demand service of computer system resources, especially CPU, memory, and I/O devices without direct access by the user. The resources are available to users through the internet via the pay-as-you-go policy. In the cloud environment, task scheduling and resource provisioning are significant problems in optimizing the system's performance. Clients increasingly rely on data storage in handling and processing of their data. This process consumes considerable amounts of energy, which hikes the operational costs substantially. In order to discourse these challenges, the paper proposes Energy-Efficient Load Balancing model based on Markov Decision Process (MDP). The EELB model is critical to a well-balanced workload distribution as well as the minimization of power consumption The proposed model leverages MDP to make informed decisions in resource allocation, ensuring optimal power utilization across virtual machines (VMs). The proposed approach is tested and compared with existing load-balancing models using various Quality of Service (QoS) metrics. In addition, an energy-efficient task allocation system is proposed to reasonably balance task scheduling and energy saving. The simulation results show that the proposed solution significantly reduce energy consumption and achieve better performance while satisfying the deadline constraint as compared to current energy-efficient scheduling methods like RC-GA, AMTS, and E-PAGA. Extensive experimentations reveal that the proposed EELB technique enhances availability by 96%, achieves 52% in energy savings, and reduces failure rates significantly compared to existing techniques and enhances throughput.