An Enhanced Virtualization of Resources for High Performance Applications in Cloud Computing Using Deep Regression Model

This paper proposes an Enhanced Virtualization of Resources (EVR) system for high performance applications in Cloud Computing. It uses a Deep Regression Model (DRM) to predict the resource requirements of an application to be deployed on the Cloud. The model takes into account various parameters like number of users, bandwidth requirements, processing time, number of I/O requests and server capability to make accurate predictions. The model is further optimized with a Genetic Algorithm, which uses mutation, crossover and selection operations to ensure the model produces a high-accuracy output. The resulting model is then used by the EVR system to decide which server nodes should be allocated to the application for best performance. The performance of the EVR system is evaluated using metrics such as Cloud query time, Cloud response time, and Cloud resource allocation accuracy. Results demonstrate that the proposed system can provide up to 73.3% more efficiency than existing approaches in Cloud virtualization.