MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641106542 A) filed by S R University on September 04, 2026, for A Blockchain-Integrated Deep Reinforcement Learning System For Dynamic, Trust-Aware, And Energy-Efficient Workload Allocation In Heterogeneous Cloud Environments.
Inventors include Dr. Ganji Ramanjaiah; and Dr. Balajee Maram.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: The present invention relates to a Blockchain-Integrated Deep Reinforcement Learning (DRL) System for Dynamic, Trust-Aware, and Energy-Efficient Workload Allocation in Heterogeneous Cloud Environments. The system dynamically allocates incoming workloads among heterogeneous physical or virtual cloud nodes by jointly considering workload requirements, resource availability, execution performance, energy consumption, and node trustworthiness. A resource monitoring module obtains current CPU, memory, storage, network, queue, and energy conditions, while a trust evaluation module determines dynamic trust scores based on historical execution performance and reliability. A Deep Q-Network (DQN) based DRL module processes the workload and cloud-state information to select a suitable cloud node using a multi-objective reward mechanism. A blockchain network validates and immutably records workload allocation, resource usage, execution status, and energy-related information. Following workload execution, performance results are fed back to update the node trust score and DRL model through experience-based learning. The integrated closed-loop mechanism enables adaptive workload allocation while improving resource utilization, reducing execution latency and energy consumption, and enhancing trust, reliability, and transparency in heterogeneous cloud environments.
Disclaimer: Curated by HT Syndication.
