The Evaluation Method towards the Application Adaptability of Internet Architecture
Zhu Mi · Chinese Journal of Computers · 2013
The development and innovation of Internet architecture have become the consensus in academia and industry.Many strategies have been proposed to fix various problems of current architecture,and even many new architectures are being considered.However,few of these new strategies or architectures have been accepted and employed in real applications.It is therefore an important and urgent task to construct a model that can be used to evaluate these strategies and architectures in response to the development requirements of Internet architecture.Most existing works used to study the competitiveness and the development trend of candidate architectures are mainly based on the economic framework,only taking into considerations of the costs and benefits to the users and ISP(or providers),while ignoring the impact of the change in the composition of application data.This paper fills this gap by proposing a method to evaluate the application adaptation capacity of the Internet architecture from the application perspective.A corresponding 2ACT(Application Adaptation CapaciTy)model is also constructed.This method constructs the model based on the service adaptability and economic adaptability of the architecture and quantifies the effects of application data composition and application deployment cost on the development of the architecture.It can be used to evaluate not only the development trend of the architecture but also the development capacity of the protocols and mechanisms within the architecture.In light of the fast growing multimedia data traffic in the Internet,this paper uses the proposed model to evaluate the two multimedia application mechanisms,In-network content caching mechanism and IP multicast protocol.The evaluation results show that these two theoretically sound mechanisms only perform well under certain situations.ISPs and customers can make their own choice based on their own application needs and our findings.