For improving, availability, scalability, reliability, flexibility, and security of cloud computing, in this research, we have presented an effective framework for accurate predictive analytics based on superior computational intelligence, deep machine learning in cloud environment. Based on a dataset from Kaggle named “Cloud Computing Performance Metrics”, we built models to achieve efficient predictions for target performance parameters such as CPU, RAM, and energy consumption. To identify the key factors, our approach incorporated advanced deep learning structures including CNNs and RNNs that enhanced the predictability of results. In the same way, we embedded benchmarking approaches for recommending preferable approaches to the usage of resources in cloud services and successful cloud operation to offer better cloud management. It is beneficial in the context of improving the performance of cloud services both in terms of inference accuracy and practical recommendations for the management of the cloud resources.