arXiv subjects

Giuseppe Siracusano

Publications and source records attributed to Giuseppe Siracusano.

21 records · Page 2Linked to original sources

arXiv2014

OSHI - Open Source Hybrid IP/SDN networking (and its emulation on Mininet and on distributed SDN testbeds)

The introduction of SDN in IP backbones requires the coexistence of regular IP forwarding and SDN based forwarding. The former is typically applied to best effort Internet traffic, the latter can be used for different types of advanced services (VPNs, Virtual Leased Lines, Traffic Engineering...). In this paper we first introduce the architecture and the services of an "hybrid" IP/SDN networking scenario. Then we describe the design and implementation of an Open Source Hybrid IP/SDN (OSHI) node. It combines Quagga for OSPF routing and Open vSwitch for OpenFlow based switching on Linux. The availability of tools for experimental validation and performance evaluation of SDN solutions is fundamental for the evolution of SDN. We provide a set of open source tools that allow to facilitate the design of hybrid IP/SDN experimental networks, their deployment on Mininet or on distributed SDN research testbeds and their test. Finally, using the provided tools, we evaluate key performance aspects of the proposed solutions. The OSHI development and test environment is available in a VirtualBox VM image that can be downloaded.

cs.NI↗

arXiv2014

Generalized Virtual Networking: an enabler for Service Centric Networking and Network Function Virtualization

In this paper we introduce the Generalized Virtual Networking (GVN) concept. GVN provides a framework to influence the routing of packets based on service level information that is carried in the packets. It is based on a protocol header inserted between the Network and Transport layers, therefore it can be seen as a layer 3.5 solution. Technically, GVN is proposed as a new transport layer protocol in the TCP/IP protocol suite. An IP router that is not GVN capable will simply process the IP destination address as usual. Similar concepts have been proposed in other works, and referred to as Service Oriented Networking, Service Centric Networking, Application Delivery Networking, but they are now generalized in the proposed GVN framework. In this respect, the GVN header is a generic container that can be adapted to serve the needs of arbitrary service level routing solutions. The GVN header can be managed by GVN capable end-hosts and applications or can be pushed/popped at the edge of a GVN capable network (like a VLAN tag). In this position paper, we show that Generalized Virtual Networking is a powerful enabler for SCN (Service Centric Networking) and NFV (Network Function Virtualization) and how it couples with the SDN (Software Defined Networking) paradigm.

cs.NI↗

arXiv2014

Controller selection in a Wireless Mesh SDN under network partitioning and merging scenarios

In this paper we consider a Wireless Mesh Network (WMN) integrating SDN principles. The Wireless Mesh Routers (WMR) are OpenFlow capable switches that can be controlled by SDN controllers, according to the wmSDN (wireless mesh SDN) architecture that we have introduced in a previous work. We consider the issue of controller selection in a scenario with intermittent connectivity. We assume that over time a single WMN can become split in two or more partitions and that separate partitions can merge into a larger one. We assume that a set of SDN controllers can potentially take control of the WMRs. At a given time only one controller should be the master of a WMR and it should be the most appropriate one according to some metric. We argue that the state of the art solutions for "master election" among distributed controllers are not suitable in a mesh networking environment, as they could easily be affected by inconsistencies. We envisage a "master selection" approach which is under the control of each WMR, and guarantees that at a given time only one controller will be master of a WMR. We designed a specific master selection procedure which is very simple in terms of the control logic to be executed in the WMR. We have implemented the proposed solution and deployed it over a network emulator (CORE) and over the combination of two physical wireless testbeds (NITOS and wiLab.t).

cs.NI↗