Open Access

ARTICLE

Monarch Butterfly Optimization for Reliable Scheduling in Cloud

B. Gomathi¹, S. T. Suganthi^2,*, Karthikeyan Krishnasamy³, J. Bhuvana⁴

1 Department of Information Technology, Hindusthan College of Engineering and Technology, Coimbatore, 641032, India
2 Department of Computer Networking, Lebanese French University, Erbil, 44001, Iraq
3 Department of Information Technology, Coimbatore Institute of Technology, Coimbatore, 641014, India
4 Department of MCA, School of Computer Science and IT, Jain (Deemed to be) University, Bangalore, 560069, India

* Corresponding Author: S. T. Suganthi. Email:

Computers, Materials & Continua 2021, 69(3), 3693-3710. https://doi.org/10.32604/cmc.2021.018159

Received 27 February 2021; Accepted 03 May 2021; Issue published 24 August 2021

Abstract

Enterprises have extensively taken on cloud computing environment since it provides on-demand virtualized cloud application resources. The scheduling of the cloud tasks is a well-recognized NP-hard problem. The Task scheduling problem is convoluted while convincing different objectives, which are dispute in nature. In this paper, Multi-Objective Improved Monarch Butterfly Optimization (MOIMBO) algorithm is applied to solve multi-objective task scheduling problems in the cloud in preparation for Pareto optimal solutions. Three different dispute objectives, such as makespan, reliability, and resource utilization, are deliberated for task scheduling problems.The Epsilon-fuzzy dominance sort method is utilized in the multi-objective domain to elect the foremost solutions from the Pareto optimal solution set. MOIMBO, together with the Self Adaptive and Greedy Strategies, have been incorporated to enrich the performance of the proposed algorithm. The capability and effectiveness of the proposed algorithm are measured with NSGA-II and MOPSO algorithms. The simulation results prompt that the proposed MOIMBO algorithm extensively diminishes the makespan, maximize the reliability, and guarantees the appropriate resource utilization when associating it with identified existing algorithms.

---

Keywords

---