Kevin Phemius

DISCO: Distributed multi-domain SDN controllers

Software-Defined Networking (SDN) is now envisioned for Wide Area Networks (WAN) and constrained overlay networks. Such networks require a resilient, scalable and easily extensible SDN control plane. In this paper, we propose DISCO, an extensible DIstributed SDN COntrol plane able to cope with the distributed and heterogeneous nature of modern overlay networks. A DISCO controller manages its own network domain and communicates with other controllers to provide end-to-end network services. This east-west communication is based on a lightweight and highly manageable control channel. We implemented DISCO on top of the Floodlight OpenFlow controller and the AMQP protocol and we evaluated it through an inter-domain topology disruption use case.

Monitoring latency with OpenFlow

Software Defined Networking, especially through protocols like OpenFlow, is becoming more and more present in networks. It aims at separating the data plane from the control plane for more network programmability, serviceability, heterogeneity and maintainability. Even if mobile applications and multimedia are often pointed at to show the demise of current network architectures, there are currently no ways to efficiently dynamically obtain the latency in an OpenFlow network to efficiently apply QoS policies. In this paper, we propose a mechanism to measure link latencies from an OpenFlow controller with high accuracy and a low footprint. We implemented it and present the performance evaluation. A monitoring packet consumes only 24 Bytes, which is 81% less than the ping utility, for an average accuracy of 99.25% compared to the ping values.

DISCO: Distributed SDN controllers in a multi-domain environment

Software-Defined Networking (SDN) is now envisioned for Wide Area Networks (WAN) and deployed constrained networks. Such networks require a resilient, scalable and easily extensible SDN control plane. In this paper, we propose DISCO, a DIstributed SDN COntrol plane able to cope with the distributed and heterogeneous nature of modern overlay networks and deployed networks. A DISCO controller manages its own network domain, communicates with other DISCO controllers to provide end-to-end network services and share aggregated network-wide information. This east-west communication is based on a lightweight and highly manageable control channel which can self-adapt to network conditions.

Failover mechanisms for distributed SDN controllers

Distributed SDN controllers have been proposed to address performance and resilience issues. While approaches for datacenters are built on strongly-consistent state sharing among controllers, others for WAN and constrained networks rely on a loosely-consistent distributed state. In this paper, we address the problem of failover for distributed SDN controllers by proposing two strategies for neighbor active controllers to take over the control of orphan OpenFlow switches: (1) a greedy incorporation and (2) a pre-partitioning among controllers. We built a prototype with distributed Floodlight controllers to evaluate these strategies. The results show that the failover duration with the greedy approach is proportional to the quantity of orphan switches while the pre-partitioning approach, introducing a very small additional control traffic, enables to react quicker in less than 200ms.

Distributed SDN for Mission-Critical Networks

Mission-critical networks now interconnect data enters, enterprise, customer sites and mobile entities. They thus must be resilient, adaptable and easily extensible. The emergence of Software-Defined Networking (SDN) protocols, which enables to decouple the control plane from the data plane, opens up new ways to architect such networks. In this paper, we propose DISCO, an extensible Distributed SDN Control plane able to cope with the distributed and heterogeneous nature of modern mission-critical networks. DISCO controllers manage their own network domain and communicate with each others to provide end-to-end network services. This inter-controller communication is based on a lightweight and highly manageable pub-sub mechanism used by agents to self-adaptively share aggregated local and network-wide information. We implemented DISCO on top of Floodlight, an Open Flow controller, and the AMQP protocol. We demonstrated how DISCOs control plane dynamically adapts to heterogeneous network topologies while being resilient enough to survive to disruptions and attacks and providing classic functionalities such as end-point migration. The experimentation results we present are organized around two use cases: inter-domain connectivity disruption and migration of a virtual machine.

Implementing OpenFlow-based resilient network services

In the OpenFlow framework, packet forwarding (data plane) and routing decisions (control plane) run on different devices. OpenFlow switches are in charge of packet forwarding, whereas a controller runs applications which decide how the packet should be handled. OpenFlow standardizes this control protocol, leaving the space for any controller implementation. In this paper we present how we developed a traffic engineering application in a multi-WAN use case with a software development approach.

Bringing SDN to the edge of tactical networks

New services put a lot of pressure on the tactical networks, always requiring more bandwidth, service differentiation and agility. In this paper we propose a framework for managing traffic in future tactical networks. Our solution is based on the Mobile Edge Cloud (MEC) architecture and its close interaction with Software Defined Networking (SDN). The whole facilitated by the spread of Software-Defined Radios (SDR) in tactical radios. We have implemented our architecture in a proof of concept and tested it with 2 tactical scenarios. Our experiments show that centralizing the intelligence in the MEC allows to guarantee the requirements of the tactical applications either by adapting the waveform parameters, or through changing the radio interface or even by reconfiguring the applications. More generally, the best decision can be seen as the optimal reaction to the wireless links variations.

NetIDE: Removing vendor lock-in in SDN

The Software-Defined Networking (SDN) paradigm allows networking hardware to be made “malleable” and remotely manageable by the so-called SDN controllers. However, the current SDN landscape is extremely fragmented. Different open and closed source controller frameworks such as Open-Daylight [1], Ryu [2], Floodlight [3], etc. exist. Porting SDN applications from one such platform to another is practically impossible and so, SDN users like network operators face a situation where they are either confined to applications working for the platform of their choice, or forced to re-implement their solutions every time they encounter a new platform.

Are We Ready to Drive Software-Defined Networks? A Comprehensive Survey on Management Tools and Techniques

In the context of the emergent Software-Defined Networking (SDN) paradigm, the attention is mostly directed to the evolution of control protocols and networking functionalities. However, network professionals also need the right tools to reach the same level—and beyond—of monitoring and control they have in traditional networks. Current SDN tools are developed on an ad hoc basis, for specific SDN frameworks, while production environments demand standard platforms and easy integration. This survey aims to foster the definition of the next generation SDN management framework by providing the readers a thorough overview of existing SDN tools and main research directions.