Publication

Abstract : Job scheduling plays a prominent part in cloud computing, and the production schedule of jobs can increase the cloud system’s effectiveness. When serving millions of users at once, cloud computing must provide all user requests with excellent performance and ensure Quality of Service (QoS). A suitable task scheduling algorithm is needed to appropriately and effectively fulfil these requests. Several methods were proposed for job scheduling in cloud computing, but the existing techniques do not provide better efficiency. The Dual Interactive Wasserstein Generative Adversarial Network Optimized with Arithmetic Optimization Algorithm Based Job Scheduling in Cloud-Based Internet of Things is proposed to overcome this issue. Primarily, the data from the Alibaba dataset is preprocessed using the Kernel Co-Relation (KC) method. The preprocessed data is given to the Dual Interactive Wasserstein Generative Adversarial Network (DIWGAN) for task forecasting in the dynamic cloud environment, and it generates scheduled tasks as output. Then the Arithmetic Optimization Algorithm (AOA) is utilized to optimize the weight parameters of the Dual Interactive Wasserstein Generative Adversarial Network. The proposed method precisely predicts the future workload and diminishes extravagant power consumption at cloud data centres. The proposed method is implemented in MATLAB. The proposed method attains lower MSE (Mean Square Error), RMSE (Root Mean Square Error), MAPE (Mean Squared Prediction Error), MAE (Mean Absolute Error), and higher results without optimization algorithm before and after normalization compared to the current approaches.