Abstract-Network Service Chaining (NSC) is a service deployment concept that promises increased flexibility and cost efficiency for future carrier networks. NSC has received considerable attention in the standardization and research communities lately. However, NSC is largely undefined in the peer-reviewed literature. In fact, a literature review reveals that the role of NSC enabling technologies is up for discussion, and so are the key research challenges lying ahead. This paper addresses these topics by motivating our research interest towards advanced dynamic NSC and detailing the main aspects to be considered in the context of carrier-grade telecommunication networks. We present design considerations and system requirements alongside use cases that illustrate the advantages of adopting NSC. We detail prominent research challenges during the typical lifecycle of a network service chain in an operational telecommunications network, including service chain description, programming, deployment, and debugging, and summarize our security considerations. We conclude this paper with an outlook on future work in this area.

No abstract

No abstract

Telecom providers struggle with low service flexibility, increasing complexity and related costs. Although "cloud" has been an active field of research, there is currently little integration between the vast networking assets and data centres of telecom providers. UNIFY considers the entire network, from home networks up to data centre, as a "unified production environment" supporting virtualization, programmability and automation and guarantee a high level of agility for network operations and for deploying new, secure and quality services, seamlessly instantiatable across the entire infrastructure. UNIFY focuses on the required enablers and will develop an automated, dynamic service creation platform, leveraging fine-granular service chaining. A service abstraction model and a proper service creation language and a global orchestrator, with novel optimization algorithms, will enable the automatic optimal placement of networking, computing and storage components across the infrastructure. New management technologies based on experience from DCs, called Service Provider DevOps, will be developed and integrated into the orchestration architecture to cope with the dynamicity of services. The applicability of a universal node based on commodity hardware will be evaluated in order to support both network functions and traditional data centre workloads, with an investigation of the need of hardware acceleration.

ocio-economic drivers, progress in information technologies, tumbling switching and computer hardware costs, and availability of open source software solutions, are creating the conditions for a change of paradigm in designing and operating telecommunication networks and service infrastructures. Network function virtualization (NFV) [1] by the European Telecommunications Standards Institute (ETSI), and software defined networking (SDN) [2] by the Open Networking Forum seem to be key technology enablers in the direction of meeting requirements such as cost reductions and new business models. NFV, on the one hand, targets at virtualizing servers and appliances that provide network functions. One of NFVs value propositions is cost optimization with the use of commercial off the shelf hardware to reduce capital expenditure. This, together with increased operational efficiency, is expected to also reduce operational expenditure. Operational efficiency is achieved by the automation of commission, configuration, resource management, etc., for even an order of magnitude higher number of managed elements than before. SDN, on the other hand, targets at breaking the vertical integration of network data and control planes to introduce (logically centralized) control plane programmability for novel networking virtualization (abstraction), simplified network (re)configuration, and policy enforcement.Today, all these promises are not yet realized. A typical example is the device located at the edge of carrier networks.Here fixed network operators require broadband network gateway (BNG) functionality for customer centric processes such as authentication, policing, etc. Technically, those gateways are physical devices hosting special network services, or software combining network control plane and data plane aspects for fulfilling the processes, sometimes even supported by hardware acceleration. Given the fact that the broadband network gateway is operated like an appliance, it is difficult to add and evolve the service offering. For example, adding a parental control function or an intrusion detection system (IDS) for a certain customer is rather cumbersome. Today, this would require the instantiation of a separate IDS application, forwarding enforcement between the network gateway and the intrusion detection, and a policy interface to configure blocking of malicious traffic. Dynamism in changing control parameters such as addresses adds additional management complexity.The NFV framework [3] promises to remedy the problems of flexible service creation by managing and orchestrating softwarized network functions into telecommunication data centers. However, considering the full lifecycle of flexible network services, e.g. fulfillment, assurance, and billing [4], one must look beyond the fulfillment phase and see that flexibility is also ensured during assurance (operation) phase.We investigate how network services can scale up/down or scale in/out to offer real elastic, pay as use services. Our main contribution is the analysis o...