Impact of chaotic initial population on the convergence of Goa-based task scheduler

• Main Authors: Shaheed, Iman Mousa, Taqi, Mustafa Kadhim, Mohammed Ali, Jamal Arkan
• Format: Conference Proceeding
• Language: English
• Subjects: Algorithms; Cloud computing; Convergence; Heuristic methods; Optimization; Performance evaluation; Scheduling; Simulation; Task scheduling
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0200055

Large-scale scientific and corporate applications have lately witnessed a high acceptance rate of cloud computing because it allows for the instant deployment of a shared pool of computing resources, including networks, storage, and servers, whenever they are needed. Every one of these programs relies on the successful culmination of multiple actions to function properly. Although it is NP-hard, task scheduling is a crucial component of the system that manages the cloud’s computing resources to guarantee QoS performance for throughput, customers regarding response time, other KPIs, and total execution time (makespan). In addition, efficient work scheduling helps cloud service providers save money on operating expenses like power and hardware. As a metaheuristic-based task scheduler, this study applies Chaotic Grasshopper Optimization Algorithm (CGOA) to the issue of efficient scheduling of tasks. CGOA used is to avoid GOA convergent early in the optimization process. The essential idea is to generate a chaos map that enlarges the potential areas of investigation and adds spice. CGOA is evaluated against the performance of GOA and PSO through extensive simulation utilizing the CloudSim toolkit simulation environment. The CGOA is shown to increase performance in task scheduling through reduced cost and time by simulation result.