Task Scheduling and Workload Forecasting Approach for Cloud Environments

As the need for data storage, processing, and advanced computing continues to rise, there is a growing demand across diverse engineering and technology sectors.the integration of cloud computing has become nearly indispensable. However, cloud-based platforms, while offering significant capabilities, are constrained by resources and require effective task distribution for balanced workload management. This task distribution is further enhanced when there is a prior estimate of the cloud workload. Consequently, both workload forecasting and task scheduling become intricate tasks that demand meticulous execution. In this paper, we introduce a methodology centered on heuristics to address task scheduling, aiming to optimize the allocation of tasks efficiently. Additionally, we introduce a machine learning-based approach for workload forecasting to provide a reliable estimate of future workloads. The performance of the task scheduling approach is evaluated with a focus on key metrics such as execution time and Central Processing Unit (CPU) usage. Simultaneously, the effectiveness of workload forecasting approach is assessed based on criteria including regression, mean absolute percentage error, mean square error, and the numbers of iterations required for training. Through this comprehensive evaluation, we aim to gauge the effectiveness of our proposed methods in enhancing the efficiency and performance of cloud computing systems.