MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202621075911 A) filed by Symbiosis International Deemed University on June 18, 2026, for System And Method For Hybrid Ai-Driven Cloud Task Scheduling.
Inventors include Sukhada Bhoyar; and Dr. Bhupesh Kumar Dewangan.
The application for the patent was published on August 07, 2026, under issue no. 32/2026.
Abstract: ABSTRACT SYSTEM AND METHOD FOR HYBRID AI-DRIVEN CLOUD TASK SCHEDULING The present invention provides a system (100) and method for load balanced task scheduling in cloud computing environments. The system comprises a cloud simulation framework (110) with a datacenter (112) and physical host (114), a virtual machine provisioning layer (120) deploying virtual machines (130) with time-shared scheduling (132), a task generation module (160) creating heterogeneous cloudlets (170), and a hybrid scheduling engine (135). The hybrid scheduling engine (135) integrates a DRL-inspired load-aware decision module (140) having a dynamic VM load map (142) and a least-loaded VM selection policy (144), with an ACO-based exploration engine (150). For each cloudlet, the system identifies the VM with minimum cumulative load and assigns the cloudlet accordingly, updating the load map in real-time. A performance metrics computation module (180) computes makespan and load balance, and a statistical validation module (190) confirms significance. [
Disclaimer: Curated by HT Syndication.
