A QoS Evolutionary Method of Cloud Service Based on User Utility Model
Sponsor
The authors wish to thank Natural Science Foundation of China under Grant NO. 61262082, 61262017 and 61462066, Natural Science Foundation of Inner Mongolia under Grand NO. 2012JQ03, 2015MS0608 and 2015MS0601, Research Program of science and technology at Universities of Inner Mongolia Autonomous Region Grant NO. NJZY008 and Inner Mongolia Science and technology innovation team of Cloud Computing and Software Engineering.
Published In
Services Computing (SCC), 2016 IEEE International Conference on
Document Type
Citation
Publication Date
9-1-2016
Abstract
The quality-of-service (QoS) is a common focus which users and service providers pay close attention to at present. For service providers, it is one of their main targets to find the optimal QoS strategy based on user preferences. However, because of the fuzziness of user preferences and the complexity of service environment, searching an optimal service strategy becomes a difficult problem. In the paper, how the QoS affects a user's satisfaction is analyzed, and then a quantitative relationship between QoS and user satisfaction is built. Based on the relationship, a user utility model of cloud service is established. In order to maximize user utility, a QoS evolutionary algorithm based on user utility model is proposed. In the algorithm, some improvement is designed to balance the contradiction between search scope and search speed in the traditional genetic algorithm. It can be seen through the experiments that the QoS optimization strategy of cloud service output by the QoS evolutionary algorithm is consistent with the target user's preferences, which can effectively enhance the cost effectiveness of service resources.
Locate the Document
DOI
10.1109/SCC.2016.79
Persistent Identifier
http://archives.pdx.edu/ds/psu/20213
Citation Details
Y. Wang, J. Zhou, J. Liu, T. Au and X. Song, "A QoS Evolutionary Method of Cloud Service Based on User Utility Model," 2016 IEEE International Conference on Services Computing (SCC), San Francisco, CA, 2016, pp. 571-576.