Premium
AirNet: An Edge‐Fabric abstraction model to manage software‐defined networks
Author(s) -
Aouadj Messaoud,
Lavinal Emmanuel,
Desprats Thierry,
Sibilla Michelle
Publication year - 2017
Publication title -
international journal of network management
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.373
H-Index - 28
eISSN - 1099-1190
pISSN - 1055-7148
DOI - 10.1002/nem.1983
Subject(s) - computer science , hypervisor , modularity (biology) , debugging , abstraction , reusability , interface (matter) , enhanced data rates for gsm evolution , distributed computing , software , abstraction layer , control (management) , software engineering , computer network , operating system , virtualization , cloud computing , telecommunications , artificial intelligence , philosophy , epistemology , bubble , maximum bubble pressure method , biology , genetics
Summary Software‐defined networking (SDN) is one of the most significant initiatives to make current networks easier to configure, test, debug, and evolve. Within an SDN ecosystem, the Northbound interface is used by operators to define policies and to program the control plane, it thus represents a major challenge. Ideally, this northbound interface should allow administrators to describe, as simply as possible, network services and their interactions, rather than specifying how and on what physical device they need to be deployed. In this article, we propose a new network control language, called AirNet, which is built on top of an abstraction model whose main feature is to provide a clear separation between edge and core network devices. Thus, logical boundaries between different types of policies will exist (control, data, and transport services), ensuring modularity and reusability of the control program. An implementation of the AirNet language has also been done, which includes in particular a hypervisor that achieves the composition of the control policies on the virtual network and their mapping on the physical infrastructure. Finally, several experiments showing promising results have been conducted on different use cases.