Quantification of Quality Expressions for Cloud Migration Decision Support

Abstract

Cloud computing is an internet-based technology, that turned in premise practice of traditional computing technology to a different approach called off-promises utilization of computing infrastructure such as storage, computer and bandwidth, in pay as you go bases. This emerging technology is being adopted by varies companies. The main reason for its fast and growing acceptance is its characteristics of being omnipresent, on demand access to a shared pool of services that can be configured, provisioned and de-provisioned with a minimum configuration effort, are a few to mention. On the other hand, cloud computing spending is growing faster than expected and a large queue of cloud computing industries are being providing the service. However, beside its vast acceptance; the technology experiences limitation and drawback from both sides’: customers and providers. For instance, interoperability, and lack of common service level agreement (SLA) management incurred difficulty in the provider selection process. On top of this, the functionality, which the large number of providers is providing, is the same, so that it is a means for an ambiguity to select the best provider of customers’ choice. To mitigate this problem, researches have been conducted and urged that it is better to choose the best cloud provider a head than a late excuse. Although some of them addressed hybrid i.e. both the personal and objective as well as a complete judgmental approach, none of them concentrate on a complete objective assessment of provider’s service, where quantification of QoS is a base. From this perspective, the underline work concentrates on quantification of cloud’s QoS that are meant used for cloud migration decision support. Thus, in this study, we have selected and proposed eighteen QoS mathematical models. The eighteen criteria are selected based on twenty-five recent research papers demand while the demand threshold is set to four. Having the established mathematical model for QoS, a case study was used, with a prototype system developed for this purpose, to assimilate the importance of the proposed model and justify its discrimination power the mathematical models. In addition to the case study, the proposed method is validated against DSRM validation method, Moody and Shakes quality framework and the four principles of Osterl et al. According to the evaluation result the proposed mathematical model can discriminate cloud providers based on the QoS they possess as well as it is justified that the proposed model is complete, integral, flexible, understandable, correct, simple and abstract.