Enhancing Cloud Computing Resource Allocation with LSTM-Based Predictive Modeling

Cloud resource demands have become more complicated as big data and AI have progressed, with characteristics such as being abrupt, arriving in bunches, and diversified. This has led to resource allocation lagging far behind the requests and an uneven resource utilization that wastes resources. Predicting when resources will be needed is done so that apps can get their hands on them right when they're needed, satisfying their dynamic resource requirements. Maintaining uniformity in service provision is challenging because of fluctuating consumer demand. This research presents a proactive approach to cloud computing resource allocation using LSTM to predict resource demands. In particular, this approach suggests an LSTM prediction method that enhances the accuracy of resource request predictions and then constructs a Round Robin resource allocation algorithm that reduces allocation latency and evenly distributes the various types of resources available on a physical machine.