Energy-Aware Cloud Monitoring Via Constrained Optimization for ServerLoad Management

R Raja Kumar1, Bethalam Madhav Varma2, Bandaru Likhitha3, Sanepalle RushiKeshava Redyy4,Peddineni Vamsi Krishna5
1Assistant Professor, Dept of Information Technology, SV College of Engineering, Tirupathi, India.
2 B.Tech , Dept of Information Technology, SV College of Engineering, Tirupathi, India.
3B.Tech , Dept of Information Technology, SV College of Engineering, Tirupathi, India.
4B.Tech , Dept of Information Technology, SV College of Engineering, Tirupathi, India.
5B.Tech , Dept of Information Technology, SV College of Engineering, Tirupathi, India.
Email:1rajakumar.r@svce.edu.in,2varmamadhav71@gmail.com,4rishireddy365@gmail.com,

3likhithabandaru2005@gmail.com,5vamsikrishna11112003@gmail.com
Corresponding Author*: R Raja Kumar.

Abstract-Energy-aware cloud monitoring tracks and optimizes the energy consumption of cloud infrastructure to improve efficiency and support sustainability goals. The existing cloud computing system aims at reducing energy consumption through constrained optimization ofserver loads. The current system employs linear programming to optimize job distribution on cloud servers, considering server capacity and power consumption, thereby decreasing computation and queue waiting times. However, the existing approach faces limitations such as partial distributions that may not fully satisfy integer constraints, reliance on fixed thresholds which reduce applicability, and insufficientintegration of energy cost in job migration decisions, which restrict comprehensive optimization and sustainability. As a proposed system to address thesechallenges, a cloud monitor is introduced that optimizes  job allocation using enhanced linear programming models incorporating workload redistribution andelasticity with energy efficiency as a primary objective. This system offers significant benefits including substantial energy savings and improved correlation between job disribution and server energy consumption,increased resource utilization, and reduced queue lengths and waiting times. The proposed approach enables dynamic load balancing by factoring in serverperformance per watt and migration costs, contributing to more sustainable and efficient cloud computing environments. Keywords: energy consumption, linear programming,cloud monitor, resource utilization, migration costs