An RNN-Based Resource Allocation Strategy in Cloud Computing

Abstract

Cloud computing has become an essential technology for businesses and individuals, providing a flexible and scalable way to store, process and access data and applications. As cloud usage continues to grow, there is a need to optimize resource allocation and minimize costs. The auction mechanism in cloud computing is important because it ensures that cloud resources are allocated to customers based on their needs and budget. Cloud providers set the initial price for the resources and customers can bid above or below this price. The auction typically lasts for a specified time, and the resources are allocated to the highest bidder at the end. This paper proposes an RNN-based resource allocation strategy for cloud computing. The proposed strategy leverages the power of recurrent neural networks to predict future resource demands of cloud applications and dynamically allocate resources accordingly. The RNN model is trained on historical data of resource usage and can capture the temporal patterns and correlations in the data. The proposed strategy is evaluated through experiments on a real-world cloud platform, and the results demonstrate that it outperforms existing resource allocation strategies regarding both resource utilization and application performance. The Proposed approach outperforms existing mechanisms in terms of enhancing the efficiency and effectiveness of cloud computing systems.