Optimization of System Performance through Ant Colony Optimization: A Novel Task Scheduling and Information Management Strategy for Time-Critical Applications

Abstract

Optimization of task scheduling and information storage/retrieval is crucial for managing resource utilization, which enhances system performance and ultimately impacts provider productivity and customer satisfaction. Efficient task scheduling aims to optimize computing time, while efficient information management focuses on maximizing memory usage. This paper presents a novel approach to task scheduling using Ant Colony Optimization (ACO) to improve time-critical objectives such as makespan and network latency, while maintaining balanced load distribution across systems. By enhancing makespan, we aim to maximize CPU utilization, and by optimizing information storage/retrieval, we target minimizing network latency. Performance across these multiple objectives is achieved by modifying the heuristic and visibility functions to guide ants toward specific solutions. The effectiveness of the proposed algorithm, Resource-Aware Load-Balancing for Time-Critical Applications (RALB-TCA), is demonstrated through implementation in the CloudSim simulation platform and benchmarking against existing techniques.