Quantitative analysis of cloud-based streaming services

Abstract

As the practitioner uses the Internet and the cloud infrastructure to facilitate online media and entertainment streaming services in large scale, it is urgent to have the quantitative analysis for the service design. The purpose of the quantitative analysis is to facilitate a high-standard service quality of such cloud-based streaming services. Naturally the service level is related to the following factors: the processing ability on the server side, the bandwidth allocated for the customer, the traffic condition of the Internet, and the processing ability of the end device on the client side. However, there is a lack of formal quantitative analysis, nor theoretical exploration, of the relationship between the service quality and these four factors. This study addresses this theoretical gap with a proposed service framework. Its queuing models for the cloud-based streaming services lead to closed form expressions for service quality of system and the corresponding computer resource required. A simulation procedure is also proposed to catch the dynamics of the presented cloud-based streaming services, estimating operation characteristics such as lag time and interruptions that a customer may experience under different environment settings. © 2013 IEEE.