The Internet has led to the creation of a digital society, where (almost) everything is connected and is accessible from anywhere. However, despite their widespread adoption, traditional IP networks are complex and very hard to manage. It is both difficult to configure the network according to predefined policies, and to reconfigure it to respond to faults, load, and changes. To make matters even more difficult, current networks are also vertically integrated: the control and data planes are bundled together. Software-defined networking (SDN) is an emerging paradigm that promises to change this state of affairs, by breaking vertical integration, separating the network's control logic from the underlying routers and switches, promoting (logical) centralization of network control, and introducing the ability to program the network. The separation of concerns, introduced between the definition of network policies, their implementation in switching hardware, and the forwarding of traffic, is key to the desired flexibility: by breaking the network control problem into tractable pieces, SDN makes it easier to create and introduce new abstractions in networking, simplifying network management and facilitating network evolution. In this paper, we present a comprehensive survey on SDN. We start by introducing the motivation for SDN, explain its main concepts and how it differs from traditional networking, its roots, and the standardization activities regarding this novel paradigm. Next, we present the key building blocks of an SDN infrastructure using a bottom-up, layered approach. We provide an in-depth analysis of the hardware infrastructure, southbound and northbound application programming interfaces (APIs), network virtualization layers, network operating systems (SDN controllers), network programming languages, and network applications. We also look at cross-layer problems such as debugging and troubleshooting. In an effort to anticipate the future evolution of this new paradigm, we discuss the main ongoing research efforts and challenges of SDN. In particular, we address the design of switches and control platforms—with a focus on aspects such as resiliency, scalability, performance, security, and dependability—as well as new opportunities for carrier transport networks and cloud providers. Last but not least, we analyze the position of SDN as a key enabler of a software-defined environment.

https://publications.uni.lu/bitstream/10993/20596/1/survey_on_SDN.pdf

Software-Defined Networking: A Comprehensive Survey

P4 is a high-level language for programming protocol-independent packet processors. P4 works in conjunction with SDN control protocols like OpenFlow. In its current form, OpenFlow explicitly specifies protocol headers on which it operates. This set has grown from 12 to 41 fields in a few years, increasing the complexity of the specification while still not providing the flexibility to add new headers. In this paper we propose P4 as a strawman proposal for how OpenFlow should evolve in the future. We have three goals: (1) Reconfigurability in the field: Programmers should be able to change the way switches process packets once they are deployed. (2) Protocol independence: Switches should not be tied to any specific network protocols. (3) Target independence: Programmers should be able to describe packet-processing functionality independently of the specifics of the underlying hardware. As an example, we describe how to use P4 to configure a switch to add a new hierarchical label.

/pdf/p4-programming-protocol-independent-packet-processors-jlla1kwsyr.pdf

P4: programming protocol-independent packet processors

The idea of programmable networks has recently re-gained considerable momentum due to the emergence of the Software-Defined Networking (SDN) paradigm. SDN, often referred to as a ''radical new idea in networking'', promises to dramatically simplify network management and enable innovation through network programmability. This paper surveys the state-of-the-art in programmable networks with an emphasis on SDN. We provide a historic perspective of programmable networks from early ideas to recent developments. Then we present the SDN architecture and the OpenFlow standard in particular, discuss current alternatives for implementation and testing of SDN-based protocols and services, examine current and future SDN applications, and explore promising research directions based on the SDN paradigm.

/pdf/a-survey-of-software-defined-networking-past-present-and-6tt309ed7r.pdf

A Survey of Software-Defined Networking: Past, Present, and Future of Programmable Networks

Software-Defined Networking (SDN) is an emerging paradigm that promises to change this state of affairs, by breaking vertical integration, separating the network's control logic from the underlying routers and switches, promoting (logical) centralization of network control, and introducing the ability to program the network. The separation of concerns introduced between the definition of network policies, their implementation in switching hardware, and the forwarding of traffic, is key to the desired flexibility: by breaking the network control problem into tractable pieces, SDN makes it easier to create and introduce new abstractions in networking, simplifying network management and facilitating network evolution. In this paper we present a comprehensive survey on SDN. We start by introducing the motivation for SDN, explain its main concepts and how it differs from traditional networking, its roots, and the standardization activities regarding this novel paradigm. Next, we present the key building blocks of an SDN infrastructure using a bottom-up, layered approach. We provide an in-depth analysis of the hardware infrastructure, southbound and northbound APIs, network virtualization layers, network operating systems (SDN controllers), network programming languages, and network applications. We also look at cross-layer problems such as debugging and troubleshooting. In an effort to anticipate the future evolution of this new paradigm, we discuss the main ongoing research efforts and challenges of SDN. In particular, we address the design of switches and control platforms -- with a focus on aspects such as resiliency, scalability, performance, security and dependability -- as well as new opportunities for carrier transport networks and cloud providers. Last but not least, we analyze the position of SDN as a key enabler of a software-defined environment.

Network virtualization is recognized as an enabling technology for the future Internet. It aims to overcome the resistance of the current Internet to architectural change. Application of this technology relies on algorithms that can instantiate virtualized networks on a substrate infrastructure, optimizing the layout for service-relevant metrics. This class of algorithms is commonly known as "Virtual Network Embedding (VNE)" algorithms. This paper presents a survey of current research in the VNE area. Based upon a novel classification scheme for VNE algorithms a taxonomy of current approaches to the VNE problem is provided and opportunities for further research are discussed.

Virtual Network Embedding: A Survey

Today diagnosing problems, deploying new services, testing protocol interactions, or validating network configurations are still largely unsolved problems for both enterprise and Internet Service Provider (ISP) networks. Due to the intrinsically distributed nature of network state, frequent timing dependencies, and sources of non-determinism involved, any change may introduce undesired effects-even the impact of a simple configuration change can be hard to predict. In this paper we show how to leverage network virtualization to improve our debugging ability: By replicating and cloning production networks and then applying the changes to the cloned network in a safe fashion. Mirror VNets thus enable troubleshooting as well as safe upgrades to both software and configuration.

/pdf/network-troubleshooting-with-mirror-vnets-1ngpa2p75b.pdf

Network troubleshooting with Mirror VNets

First and second controllers implemented on computing equipment may be used to control switches in a network. The switches may forward network packets between end hosts. The second controller may identify first and second redundant partitions of switches in the network that are each coupled to all of the end hosts. The first controller may instruct the first partition to install software while the second partition forwards network traffic and may instruct the second partition to install software while the first partition forwards network traffic. The first controller may install the software while the second controller is active and the second controller may install the software while the first controller is active. In this way, the switches and controllers may be provided with an uninterrupted software upgrade and packets may be forwarded between end hosts during the software upgrade without introducing packet loss or other noticeable reductions in network performance.

Systems and methods for performing uninterrupted network upgrades with controllers

Multimedia applications have recently been challenging existing mobile access networks and are raising the bar for next generation mobile networks (NGMN), both in terms of network traffic as well as in the expectations of end users. At the same time, the network and server landscape sees changes due to the advent of virtualization and split-architecture networks like OpenFlow. In this heterogeneous environment, quantitative measurement and prediction of the user’s Quality of Experience (QoE) require testbeds capable of studying these effects combined as well as in isolation, in a controlled and reproducible manner. In this paper, we present the design and architecture of QoE-Lab, a multi-purpose heterogeneous testbed that supports a variety of networking conditions to evaluate QoE in future Internet scenarios. QoE-Lab includes 1) NGMN networks, 2) access/backbone network emulation, and 3) virtualization. It provides services like traffic generation, topology emulation and high-precision cross-layer monitoring. The experiments are provisioned, orchestrated and analyzed by a tool called ExpAuto. We analyze the experimental results for relationships between network performance and user perception. We report initial results from our studies, qualitatively indicating that each of the discussed components has a significant impact on the QoE.

/pdf/qoe-lab-towards-evaluating-quality-of-experience-for-future-28lq0pqyqk.pdf

QoE-Lab: Towards Evaluating Quality of Experience for Future Internet Conditions

SUMMARY The data rates provisioned by broadband Internet access connections continue to fall short of the requirements posed by emerging applications. However, the potential of statistical multiplexing of the last mile broadband connections remains unexploited even as the average utilisation of these connections remains low. Despite recent work in this area, two key questions remain unanswered: (a) what is the attainable benefit of broadband access sharing? and (b) how much of this benefit is realisable given real-world constraints? In this work we quantify the attainable benefit of a multihomed broadband access environment by proposing and evaluating several flow-based access sharing policies using a custom flow-based simulator. We then analyse how much of the performance benefit is lost due to real-world constraints by migrating from simulations to a test-lab environment employing a wireless network. Our results show that in today’s broadband Internet access scenarios, a significant reduction in download times (up to a factor of 3) is achievable. Copyright © 2009 John Wiley & Sons, Ltd.

https://www.net.t-labs.tu-berlin.de/papers/MWFV-PBFRME-09.pdf

Potential Benefit of Flow-based Routing in Multihomed Environments

Designing infrastructures for automatically conducting controlled, reproducible Internet experiments poses substantial challenges. These include modelling the Internet structure in a useful way, and providing sufficciently Internet-like traffic. In this paper we describe a multi-purpose experimental platform, called Routerlab, enableing complex Internet measurement experimentation. Departing from the approach taken by many popular testbeds, the Routerlab is a router centric, customizable testbed that reflects the macroscopic structure of the Internet, with different clouds and tier levels and simulation of differing access properties via network emulation. This, in combination with a flexible traffic generation framework, enables us to conduct Internet-level experiments on a small scale, within the borders of our server room. Routerlab?s management platform called Labtool supports heterogeneous hardware landscapes and user groups (e.g., labcourse student teams, individual researchers working on differing layers, developers). To accomodate the span of differing requirements of these user groups, it offers both virtualized and bare-metal, non-virtualized resources and can support versioned, controlled changes to the physical topology.

/pdf/the-routerlab-emulating-internet-characteristics-in-a-room-31aqhnzi65.pdf

Andreas Wundsam

Papers

Network troubleshooting with Mirror VNets

Systems and methods for performing uninterrupted network upgrades with controllers

QoE-Lab: Towards Evaluating Quality of Experience for Future Internet Conditions

Potential Benefit of Flow-based Routing in Multihomed Environments

The Routerlab - Emulating Internet Characteristics in a Room