MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641105420 A) filed by Ayyappa Srinivasan M G; P. Petchiammal; Dr. Vinod Kumari; Ms. Shruti Jindal; Ms. Anju Saini; Dr. Shaifali; Dr. Sumit Kumar; Dr. Jitendra Kumar Seth; Ms. Nivetha S; and Dr. S Vinayagapriya on September 01, 2026, for Energy-Aware Virtual Machine Placement Using Multi-Objective Optimization.
Inventors include Ayyappa Srinivasan M G; P. Petchiammal; Dr. Vinod Kumari; Ms. Shruti Jindal; Ms. Anju Saini; Dr. Shaifali; Dr. Sumit Kumar; Dr. Jitendra Kumar Seth; Ms. Nivetha S; and Dr. S Vinayagapriya.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: [06] Virtual machine placement is a central resource-management problem in infrastructure-as-a-service cloud computing because a placement decision simultaneously determines server activation, resource utilization, energy demand and the likelihood of service degradation. Strategies that minimize only the number of active servers often create high-utilization hotspots, whereas conservative strategies protect quality of service but leave more machines powered on. This work presents Pareto-Guided Energy-Aware Virtual Machine Placement (PGE-VMP), a multi-objective placement framework that jointly considers estimated server energy, service-level-agreement risk, multi-resource wastage and the number of active physical machines. The method uses feasibility-first placement, non-dominated comparison of candidate mappings, diversity-preserving selection, energy-biased local changes and a knee-point decision rule for choosing a balanced deployment from the final trade-off set. A deterministic Python simulation model with heterogeneous servers and workloads from 100 to 500 virtual machines is used for manuscript-level evaluation. Across the tested scales, the proposed method reduces the average SLA-risk proxy by about 25.5% compared with an energy-aware best-fit baseline while keeping average modeled energy within approximately one percent of the energy-oriented multi-objective baseline. It also reduces hotspot counts substantially, illustrating why energy cannot be optimized in isolation. The findings support multi-objective VM placement as a practical strategy for greener cloud resource management under explicit performance constraints. Accompanied Drawing [FIG. 1] [FIG. 2] [FIG. 3] [FIG. 4] [FIG. 5] [FIG. 6] [FIG. 7] [FIG. 8]
Disclaimer: Curated by HT Syndication.
