Book Description

This thesis presents the results of a study focused around the design and development of networking techniques aimed at the deployment and support of advanced services in the future Internet. After many years of constant evolution, the Internet has approached a historic inflection point where mobile platforms, applications and services are poised to replace the fixed-host/server model that has dominated the Internet since its inception. Driven by the strikingly different Internet population of mobile devices and services, new fundamental communication abstractions are required and the current IP based Internet fails to meet their requirement in a satisfying fashion. A top-down analysis of the requirements of such future mobile Internet services is provided, motivating a comprehensive set of solutions needed to meet them. Moreover, starting from the recognition that new core technologies will be a core enabling factor of the previously described evolution, driven by advances such as increased computing power and storage, as well as the trend towards software-based programmability and virtualization. This thesis not only aims to describe why such solutions are required, but also develops a bottom-up analysis of how these new technological advances could be employed to address the new requirements. The first chapter of the thesis introduces the reader to the fundamental issues at stake, discussing the central architectural concept of Named-Object based networking and the power that lies behind it. Looking at the different architectures presented over the years, a set of fundamental abstractions are defined, providing a comprehensive analysis of their properties and how they could be met. This study leads to the presentation of the MobilityFirst architecture in which the ''narrow waist'' of the protocol stack is based on Named-Objects which enable a broad range of capabilities in the network. This is followed up with a specific set of network service APIs that provide full access to the proposed abstractions supported by MobilityFirst. Using performance benchmarks and the implementation of representative use cases it is shown that the abstractions enabled by the new API are flexible and can enable efficient and robust versions of present and future applications. The second chapter of the thesis then moves to the set of services that will be required by the future mobile Internet and that due to different shortcomings are hardly supported by the current TCP/IP Internet architecture. These include: i) Multicast services, ii) Content services, iii) In-network compute, and finally iv) Context services. For each of these services, appropriate abstractions enabled by the Named-Object architecture are presented and a use case based prototype evaluation is provided. The results show the feasibility of providing a broad range of services with good performance and reasonable protocol overhead. Starting from the above abstractions analysis and the newly introduced services, the third chapter of the work, focuses on how such new services are made available to the end-users of the network. Considering first the expected requirements for such systems, a new transport layer service is presented. The new designed protocol can seamlessly support a set of distinctive features based on use of names and in-network reliability techniques. Using the developed prototype components, experimental results show that for a few representative scenarios including mobile data delivery, web content retrieval, and disconnected/late binding service, the new systems can be exploited to reduce the impact of complex operations improving performance for the end users of the network. The fourth chapter analyzes how advanced cloud services can be supported in the proposed Named-Object architecture. In particular, the concept of naming is extended to natively support virtual network identifiers. It is shown that the virtual network capability can be designed by introducing the concept of a ''Virtual Network Identifier (VNID)'' which is managed as a Named-Object. Further, the design supports the concept of Application Specific Routing (ASR) which enables network routing decisions to be made with awareness of application parameters such as cloud server workload. Experimental results show that the new framework provides a clean and simple logic for defining and managing virtual networks while limiting the performance impact produced by the additional overhead generated by running such system. Moreover, the potential of ASR is demonstrated through a based cloud service use case deployment. The last chapter of the thesis aims to bring together the whole study and provide considerations on how the different components presented could be merged into a single end-to-end realization. The designed elements are used in combination to present an overview of how they could all be joined into a single experimental platform ready to be employed in various deployment scenarios. Specific prototyping details are given for several scenarios including advanced computing and context-aware services and how these have been deployed on a nation wide testbed.