Optimising resource allocation in a power and cost-aware multi-cloud environment using Multi-Objective Antlion Optimisation algorithm

With the proliferation of cloud computing and the necessity to work across multiple clouds, there is a strong need to schedule tasks to cloud resources effectively by factoring in the constraints of makespan, cost, and power utilisation.In this paper, a meta-heuristic optimisation algorithm (Quasi-o...

Full description

Saved in:
Bibliographic Details
Published in2022 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM) pp. 24 - 31
Main Authors Kumar, Prithvi Anil, Sar, Spandan, Raj, Surya, M K, Varun, Phalachandra, H L, Auradkar, Prafullata K
Format Conference Proceeding
LanguageEnglish
Published IEEE 12.12.2022
Subjects
Cloud computing
CloudSim
Containers
Costs
Meta-heuristic algorithm
Minimax techniques
Multi-cloud
Multi-objective Antlion Optimisation
Multi-objective optimisation
Scheduling algorithms
Training
Virtual machining
Online AccessGet full text

Cover

More Information
Summary:With the proliferation of cloud computing and the necessity to work across multiple clouds, there is a strong need to schedule tasks to cloud resources effectively by factoring in the constraints of makespan, cost, and power utilisation.In this paper, a meta-heuristic optimisation algorithm (Quasi-oppositional Multi-objective Antlion Optimizer based on Differential Evolution) is utilised to allocate a set of cloud tasks to heterogeneous containers and simultaneously allocate these containers to virtual machines spanning across multiple Cloud Service Providers. Experiments conducted using the CloudSim framework on two cloud task workloads show that the MOALO algorithm performs competitively in terms of makespan and significantly better in cost and power utilisation when compared to scheduling algorithms like Min-Min and Max-Min.
DOI:10.1109/CCEM57073.2022.00012