Toward A Performing Resource Provisioning Model for Hybrid Cloud

Toward A Performing Resource Provisioning Model for Hybrid Cloud

Mohammed Rebbah (University of Mascara, Mascara, Algeria), Yahya Slimani (Department, ISAMM, University of Manouba, Tunisa, Tunisia), Mohammed Debakla (University of Mustapha Stambouli, Mascara, Algeria) and Omar Smail (University of Mustapha Stambouli, Mascara, Algeria)
Copyright: © 2018 |Pages: 28
DOI: 10.4018/IJGHPC.2018100102


This article describes how the idea of a hybrid cloud comes from the coupling of public and private clouds to more efficiently address user requirements. This article addresses the problem of resource provisioning in hybrid cloud. This article is mainly concerned about optimizing the resources provisioning task through the reduction of the tasks completion time together with minimal cost and more reliable services. Two steps are considered in the proposed model, which are brokering and scheduling. In the brokering strategy, this article formalizes the problem as a minimization problem of the completion time as the objective function, under cost and service reliability constraints. The scheduling strategy contains two phases: (i) use the balanced k-means method to classify the submitted tasks and, (ii) perform a minimum assignment using the Hungarian algorithm. The proposed model is evaluated within the simulation framework CloudSim. Experimental results demonstrate that the provisioning model significantly reduces both the response time and the slowdown of user's requests for different scheduling algorithms.
Article Preview

1. Introduction

The term Cloud Computing usually refers to online delivery and consumption model for business and customer services. Clouds provide a virtualized platform for users to create and manage the software stack from the operating system to the applications. This particular type of cloud is known as an Infrastructure-as-a-Service (IaaS) cloud. The customizability, complete control over the software stack, and on-demand access to IaaS cloud make them an attractive solution to the problem of dynamically extending the resources of a static site to adjust to changes in demand (Marshall, Keahey, Freeman, 2010). In Cloud computing (Buyya, Yeo, Venugopal, Broberg, Brandic, 2009), resources can be either externally owned (public cloud), or internally owned (private cloud), the former being offered by Cloud providers. Public clouds offer access to external users who are typically billed on a pay-as-you-use economic model. Resource size is dynamic, growing by way of on-demand creation of the resources of the desired type (e.g., virtual machines VMs), this kind of dynamic sizing being supported by virtualization technologies that enable dynamic creation, migration, and destruction of resources. IaaS can customize and configure VMs-based on application demands providing massive scalability, high reliability and performance (for example, the Amazon Elastic Compute Cloud (EC2) (Amazon, n.d.a)), as well as virtual storage (for example, the Amazon Simple Storage Service (S3) (Amazon, n.d.b)).

Utilizing public Cloud services along with local resources (e.g., private Cloud) to support Hybrid Clouds (Sotomayor, Montero, Llorente, Foster, 2009A), is one of the most widely used Cloud computing model. Hybrid Cloud platforms help scientists and businesses leverage the scalability and cost effectiveness of the public Cloud by paying only for IT resources consumed (server, connectivity, storage) while delivering the levels of performance and control available in private Cloud environments without changing their underlying IT setup (He & Wang, 2015). However, efficient policies to integrate public and private Clouds remain a challenge (Javadi, Abawajy & Sinnott, 2012).

Complete Article List

Search this Journal:
Open Access Articles: Forthcoming
Volume 11: 4 Issues (2019): 1 Released, 3 Forthcoming
Volume 10: 4 Issues (2018)
Volume 9: 4 Issues (2017)
Volume 8: 4 Issues (2016)
Volume 7: 4 Issues (2015)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing