Optimization: data-driven management using deep learning in cloud computing

The data-driven framework is one of the most demanded managements of cloud computing (CC). This study was focused on optimization and deep learning (DL) for the CC framework. It requires different types of frameworks' data or resources for executing various services on the cloud framework. We propose a model-based data-driven framework that explores data-driven management in terms of CC. This research is an optimal selection of CC recovery when the uncertainty of cloud networks can be based on time constraints and objective functions. We consider the emergence of CC data centers such as Amazon Web Services (AWS) and Wikipedia. These data centers can bring a considerable risk to the uncertainty of the data quality in real-time demand from organizational CC. In the present pandemic situation, data centers perform various degradation of the quality of data because the allocation of resources has increased the input of data size, especially in cloud workload. So, we use an Artificial Neural Network (ANN) model to perform a time allocation and elasticity for managing the CC workload and their management. Real-time workload resilience, trace troubleshooting, probability, and availability are implemented to analyze the data quality for better performance, which leads to overhead on the cloud platform performance.