Software defined networking (SDN) aims to provide simplified network design, operation, and management using a decoupled control plane. However, its centralized control and global network knowledge present scalability and reliability issues, which makes SDN deployment very challenging. In this paper, we propose and evaluate a hybrid switch with partial delegation of basic bridging and new cooperative mechanisms between controller and switches. This delegation offloads the SDN controllers while maintaining the capability to install forwarding rules on the switches. In this way, we take full advantage of hybrid switches in addition to using them as backwards compatible equipment, which interoperates with traditional switches.We validate this proposal by implementing a hybrid OpenFlow switch on an open source software switch as a proof of concept. Scalability and path setup delay are improved with respect to traditional centralized SDN solutions, because of the reduction in controller load and, in turn, because of the reduced traffic between switches and controller. Our cooperative mechanisms focus on recovering failures, obtaining the best performance of all approaches on higher loads, and providing a good trade‐off between controller based and traditional distributed approaches.

Software-Defined Networking (SDN) is a pillar of next-generation networks. Implementing SDN requires the establishment of a decoupled control communication, which might be installed either as an out-of-band or in-band network. While the benefits of in-band control networks seem apparent, no standard protocol exists and most of setups are based on ad-hoc solutions. This article defines Amaru, a protocol that provides plug&play resilient in-band control for SDN with low-complexity and high scalability. Amaru follows an exploration mechanism to find all possible paths between the controller and any node of the network, which drastically reduces convergence time and exchanged messages, while increasing robustness. Routing is based on masked MAC addresses, which also simplifies routing tables, minimizing the number of entries to one per path, independently of the network size. We evaluated Amaru with three different implementations and diverse types of networks and failures, and obtained excellent results, providing almost on-the-fly rerouting and low recovery time. INDEX TERMS SDN, OpenFlow, in-band control, resilient networks, path exploration.

The Software-Defined Networking (SDN) architecture decouples the control plane from the data plane, but it does not explicitly state where the control should be located. This article analyses the benefits of maintaining the control as close as possible to the data plane, instead of the more traditional centralised control plane approach. To this purpose, it delves into the study of ARP-P4, a hybrid software switch defined by using the P4 language to facilitate its future use and deployment in P4 targets. Its hybrid properties come from supporting two complementary different ways of establishing paths: a centralised SDN approach based on P4-Runtime and a traditional distributed approach based on the ARP-Path protocol that obtains a similar performance to centralised solutions based on Equal Cost Multi-Path (ECMP) and Dijkstra. The results show the feasibility of hybrid devices that combine different forwarding paradigms without losing performance with respect to well-known solutions such as ECMP, and how their combined use can lead to enhance and scale communication networks.

No abstract

Recently, two technologies have emerged to provide advanced programmability in Software-Defined Networking (SDN) environments, namely P4 and XDP. At the same time, the Internet of Things (IoT) represents a pillar of future 6G networks, which will be also sustained by SDN. In this regard, there is a need to analyze the suitability of P4 and XDP for IoT. In this article, we aim to compare both technologies to help future research efforts in the field. For this purpose, we evaluate both technologies by implementing diverse use cases, assessing their performance and providing a quick qualitative overview. All tests and design scenarios are publicly available in GitHub to guarantee replication and serve as initial steps for researchers that want to initiate in the field. Results illustrate that currently XDP is the best option for constrained IoT devices, showing lower latency times, half the CPU usage, and reduced memory in comparison with P4. However, development of P4 programs is more straightforward and the amount of code lines is more similar regardless of the scenario. Additionally, P4 has a lot of potential in IoT if a special effort is made to improve the most common software target, BMv2.