Energy-aware Scheduling of Tasks in Cloud Computing.

Abstract

Cloud computing Infrastructures have been created to facilitate consumer access to various services through the Internet. Massive energy consumption by data centers that host Cloud applications result in high carbon footprints to the environment. Therefore, there is a need for ways to reduce the energy consumption. These aspects are reduced by efficiently task scheduling within the deadline respect and providing the resources according to the user request. We discuss energy usage, execution time, and SLA violations in virtualized cloud data centers in this study. For effective scheduling, the suggested approach is predicated on job categorization and thresholds. Tasks having lengthy execution durations are preprocessed in the first stage by being placed in different lists. The following stage involves classifying tasks according to the resources required. Finally, Genetic Algorithm is used to select the best schedules. To represent the dynamic nature of the cloud environment and to offer a scheduling solution that is nearly optimum and decrease energy consumption, execution time and SLA violation, an adaptive Genetic Algorithm is developed. By the use of cloud infrastructure simulation and a series of performance and quality assessment experiments, the suggested model is verified in this setting. Results show that the suggested method improves performance by reducing execution time, energy usage, and SLA violations.