A MULTILEVEL PRINCIPAL COMPONENT ANALYSIS BASED QOS AWARE SERVICE DISCOVERY AND RANKING FRAMEWORK IN MULTI-CLOUD ENVIRONMENT

摘要

With the rapid increase in the utilization of the cloud services, various cloud service providers are keeping their efforts in the design and development of the Quality of Service (QoS) aware composite services that satisfy the user preferences. QoS aware cloud service discovery and selection is considered as an NP-hard problem due to the existence of similar cloud services in different cloud environments. Existing cloud service selection mechanisms adopt the procedure of calculating the weighted summation of the QoS attributes to select cloud services. But due to the lack of correlation between the QoS preferences of the cloud service, these approaches may produce inaccurate results. In this paper, a multilevel principal component analysis (PCA) based service selection mechanism is proposed to discover and rank the services based on the user preferences in a multi-cloud environment. Modified PCA based service agent is deployed to select the services on analyzing the QoS correlations if each service. Finally, the experimental results show that our proposed mechanism outperforms the existing service selection techniques in terms of computation time and reduction of discovery overhead.