Open Access

ARTICLE

A Load-Fairness Prioritization-Based Matching Technique for Cloud Task Scheduling and Resource Allocation

by Abdulaziz Alhubaishy^1,*, Abdulmajeed Aljuhani²

1 College of Computing and Informatics, Saudi Electronic University, Riyadh, 11673, Saudi Arabia
2 College of Computer Science and Engineering, Taibah University, Medina, 41411, Saudi Arabia

* Corresponding Author: Abdulaziz Alhubaishy. Email:

Computer Systems Science and Engineering 2023, 45(3), 2461-2481. https://doi.org/10.32604/csse.2023.032217

Received 10 May 2022; Accepted 17 June 2022; Issue published 21 December 2022

Abstract

In a cloud environment, consumers search for the best service provider that accomplishes the required tasks based on a set of criteria such as completion time and cost. On the other hand, Cloud Service Providers (CSPs) seek to maximize their profits by attracting and serving more consumers based on their resource capabilities. The literature has discussed the problem by considering either consumers’ needs or CSPs’ capabilities. A problem resides in the lack of explicit models that combine preferences of consumers with the capabilities of CSPs to provide a unified process for resource allocation and task scheduling in a more efficient way. The paper proposes a model that adopts a Multi-Criteria Decision Making (MCDM) method, called Analytic Hierarchy Process (AHP), to acquire the information of consumers’ preferences and service providers’ capabilities to prioritize both tasks and resources. The model also provides a matching technique to assign each task to the best resource of a CSP while preserves the fairness of scheduling more tasks for resources with higher capabilities. Our experimental results prove the feasibility of the proposed model for prioritizing hundreds of tasks/services and CSPs based on a defined set of criteria, and matching each set of tasks/services to the best CSPS.

---

Keywords

---