MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202621081998 A) filed by Symbiosis International Deemed University on July 02, 2026, for System And Method For Dynamic Cloud Resource Allocation Using Lstm And Deep Reinforcement Learning.
Inventors include Rushi Parhad; and Dr. Bhupesh Kumar Dewangan.
The application for the patent was published on August 28, 2026, under issue no. 35/2026.
Abstract: ABSTRACT SYSTEM AND METHOD FOR DYNAMIC CLOUD RESOURCE ALLOCATION USING LSTM AND DEEP REINFORCEMENT LEARNING The present invention discloses a system (100) and method for dynamic resource allocation in cloud computing environments. The system comprises a workload generator module (110) for producing time-series workload data, an LSTM prediction module (120) employing a Long Short-Term Memory neural network with input gate (121), forget gate (122), output gate (123), and memory cell (124) for predicting future workload demands, a data interface layer (130) for inter-technology communication, a resource allocation engine (140) implementing a deep reinforcement learning inspired state-action-reward policy for adaptive virtual machine scaling, a cloud simulation environment (150) with modeled datacenters (151), hosts (152), virtual machines (153), and cloudlets (154), a performance metrics module (160), and an output and visualization module (170). The system achieves approximately 40 percent reduction in response time compared to static allocation while maintaining equivalent execution time, demonstrating effective integration of predictive deep learning with intelligent resource management. [
Disclaimer: Curated by HT Syndication.
