Quality of service model for software as a service in cloud computing from users' and providers' perspectives

• Main Author: Khanjani, Atieh
• Format: Thesis
• Language: English
• Published: 2015
• Online Access: http://psasir.upm.edu.my/id/eprint/57127/1/FSKTM%202015%2015RR.pdf; http://psasir.upm.edu.my/id/eprint/57127/
• Institution: Universiti Putra Malaysia

Software as a Service (SaaS) is one of the main service models in cloud computing which enables the application to run on the cloud by eliminating the installation on the personal computer at the client side. Quality of Service (QoS) is a crucial factor for the success of cloud services especially in terms of SaaS, so that if it is not delivered as expected, it might blemish the provider's reputation. In this thesis, we address the problem of the lack of QoS model for SaaS to cover more QoS attributes compared to other existing models and their definitions to be referred as reference model which are useful in both users' and providers' perspectives. There is a high demand for creating a quality model for SaaS since conventional frameworks cannot effectively support specific quality aspects of SaaS such as scalability and reusability. Even though there are some studies that have been performed regarding the QoS models for SaaS but they considered only a few attributes and still many aspects are left. Besides, the users might not have sufficient knowledge and experience of what they want and not be able to clarify their requirements very well. Therefore, the QoS consideration should be from both service users' and providers' perspectives to be more effective. In this research, QoS attributes for SaaS cloud services from both users' and providers' perspectives are presented, defined and categorized. A quality model for SaaS called SaaS-QoS model as a reference model to be useful in both perspectives, also was proposed. First, a set of 29 QoS attributes for SaaS cloud from the literature consisted of QoS attributes specific for SaaS quality models and in overall cloud computing, was obtained. Then, a survey conducted through experts in industry, academician and researchers to measure the acceptability of the attributes using purposive sampling technique. Based on the result of the survey, 32 attributes under 5 categories were determined as QoS attributes for SaaS inspired by Service Measurement Index (SMI) framework and the SaaS-QoS model was then proposed. The SaaS-QoS model was evaluated from both user and provider perspectives through performing two surveys by SaaS providers and SaaS users. The results of evaluation part indicated that the categories and attributes assigned are highly associated and relevant. More than that, the SaaS-QoS model is practical and applicable enough for SaaS from both users' and providers' perspectives. This research has given a wider view of QoS attributes to both users and providers as a reference model and also to add to the body of knowledge and practitioners.