Intellectual Property India Publishes Patent Application for Reinforcement Learning Based Adaptive Load Balancing System And Method For Cloud Computing Environments

MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202621075906 A) filed by Symbiosis International Deemed University on June 18, 2026, for Reinforcement Learning Based Adaptive Load Balancing System And Method For Cloud Computing Environments.

Inventors include Aarya Jawanjal; and Dr. Bhupesh Kumar Dewangan.

The application for the patent was published on August 07, 2026, under issue no. 32/2026.

Abstract: ABSTRACT REINFORCEMENT LEARNING BASED ADAPTIVE LOAD BALANCING SYSTEM AND METHOD FOR CLOUD COMPUTING ENVIRONMENTS The present invention discloses a reinforcement learning based adaptive load balancing system (100) and method for cloud computing environments. The system (100) comprises a cloud simulation environment (110) hosting a datacenter (120) with physical hosts (130) and a plurality of virtual machines (140). A task generation module (150) generates computational tasks with heterogeneous workloads. A reinforcement learning broker agent (160) employs Q- Learning with an epsilon-greedy exploration policy to dynamically allocate tasks to optimal virtual machines (140) based on real-time system state parameters monitored by a state observation module (170) including CPU load, memory utilization, and queue length. A reward computation module (180) evaluates allocation decisions using a reward function R = -(Execution Time + alpha x Load Imbalance), and a Q-table update module (190) refines the decision policy iteratively. [

Disclaimer: Curated by HT Syndication.