Optimal Cloud-Path Selection in Mobile Cloud Offloading Systems Based on QoS Criteria

Optimal Cloud-Path Selection in Mobile Cloud Offloading Systems Based on QoS Criteria

Huaming Wu (Free University of Berlin, Berlin, Germany), Qiushi Wang (Free University of Berlin, Berlin, Germany), and Katinka Wolter (Free University of Berlin, Berlin, Germany)
Copyright: © 2013 |Pages: 18
DOI: 10.4018/ijghpc.2013100103
OnDemand PDF Download:
No Current Special Offers


Recently, there emerge a variety of clouds in sky and thus, several similar cloud services (from different cloud venders) can be provided to a mobile end device. The goal of cloud-path selection is to find an optimal cloud-path pair between the mobile device and a cloud among a certain class of clouds that provide the same service, in order to carry out the offloaded computation tasks. It is easy to choose the optimal cloud-path to save execution time incurred by offloading program to cloud when considering only one factor. However, there are many Quality of Service (QoS)-based criteria such as performance, bandwidth, financial, security and availability that need to be considered when making final decisions. In this paper, a multiple criteria decision analysis approach based on the analytic hierarchy process (AHP) and the technique for order preference by similarity to ideal solution (TOPSIS) in a fuzzy environment is proposed to decide which cloud is the most suitable one for offloading. The AHP is used to determine the weights of the criteria for cloud-path selection, while fuzzy TOPSIS is to obtain the final ranking of alternative clouds. The numerical analysis is performed to evaluate the model. Furthermore, a method based on historical data of the mobile device’s experiences is used to evaluate the importance weights of the alternative cloud service, when it is challenge to measure and acquire the parameters of criteria timely in practical systems.
Article Preview

Literature Review

Cloud service selection is a highly significant research issue but it has not been fully investigated and little literature has been published in this area since cloud computing itself is still in its early stages. In this section, we give a brief overview of the related framework in cloud service selection.

The diversity in cloud computing offering makes it difficult to compare one cloud service against others. To help cloud users in selecting a cloud provider, CloudCmp (Li, Yang, Kandula & Zhang, 2010a, 2010b) has been proposed to compare the performance of public cloud services such as Amazon EC2, Windows Azure and Google AppEngine. A set of benchmarking tools are used in CloudCmp to compare the common services (such as elastic computing cluster, persistent storage, intra-cloud and wide area network) and the benchmarking results are then used to predict the performance and costs of application when deployed on a cloud provider.

CloudRank (Zheng, Zhang & Lyu, 2010) is a collaborative QoS-driven ranking framework for cloud components to predict the quality ranking of cloud components without requiring additional real-world component invocations from the intended user. By taking advantage of the past component usage experiences of different component users, it identifies and aggregates the preferences between pair of components to produce a ranking of the components through a proposed greed method.

Multi-Criteria Comparison Method for Cloud Computing ((MC2)2) (Menzel, Schönherr & Tai, 2011) offers a multi-criteria-based decision framework that can be applied to cloud computing scenarios. (MC2)2 framework aims to choose the most suitable one when filtering out all infeasible alternatives by evaluating and ranking candidate cloud services using multiple criteria derived from a comprehensive criteria catalog. As a recommendation multi-criteria decision making process, the analytic network process (ANP) employs pair-wise comparisons and normalization to assign values to quantitative and qualitative criteria on a ratio scale.

Complete Article List

Search this Journal:
Volume 15: 2 Issues (2023): 1 Released, 1 Forthcoming
Volume 14: 6 Issues (2022): 1 Released, 5 Forthcoming
Volume 13: 4 Issues (2021)
Volume 12: 4 Issues (2020)
Volume 11: 4 Issues (2019)
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