Intellectual Property India Publishes Patent Application for A Method For Dynamic Load Balancing In Cloud Computing Using Meta-Reinforcement Learning

MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202641062969 A) filed by Dr M Shiva Rama Krishna; Dr D Khasim Vali; Potu Bharath; Dr Madduri Venkateswarlu; Dr Y. Madhavilatha; Karrolla Bharath; Ms. Amulya P; Dr Rajkumar Kalimuthu; and Dr. Sai Babu Veesam on May 18, 2026, for A Method For Dynamic Load Balancing In Cloud Computing Using Meta-Reinforcement Learning.

Inventors include Dr M Shiva Rama Krishna; Dr D Khasim Vali; Potu Bharath; Dr Madduri Venkateswarlu; Dr Y. Madhavilatha; Karrolla Bharath; Ms. Amulya P; Dr Rajkumar Kalimuthu; and Dr. Sai Babu Veesam.

The application for the patent was published on June 12, 2026, under issue no. 24/2026.

Abstract: The present invention relates to a method for dynamic load balancing in cloud computing using Meta-Reinforcement Learning for adaptive task scheduling, intelligent workload prediction, and optimized resource allocation in distributed cloud environments. The method comprises collecting workload data from a plurality of virtual machines operating within a cloud infrastructure and predicting workload conditions using a hybrid deep learning architecture integrating Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN). The predicted workload values are utilized to classify virtual machines into overloaded virtual machines and underloaded virtual machines. A Meta-Reinforcement Learning framework integrated with a Hybrid Lyrebird Falcon Optimization engine dynamically generates adaptive scheduling policies for workload redistribution and task allocation. The method continuously optimizes Quality-of-Service parameters including makespan, energy consumption, balanced CPU utilization, scalability efficiency, resource utilization, and task failure rate. The Meta-Reinforcement Learning framework continuously updates scheduling decisions based on workload fluctuations, historical scheduling outcomes, and real-time resource availability, thereby improving cloud system scalability, computational efficiency, workload balancing, and energy optimization. The invention provides an intelligent and adaptive cloud resource management framework capable of efficient workload distribution in heterogeneous and dynamic cloud computing environments.

Disclaimer: Curated by HT Syndication.