RINA-based virtual networking solution for distributed VNFs: Prototype and Benchmarking

Author

Fernández, Carolina

Giménez, Sergio

Grasa, Eduard

Siddiqui, Shuaib

Publication date

2022-07-01



Abstract

The decoupling of the network functions from the hardware they run on has become a common practice nowadays. Complex network services made up of multiple Virtual Network Functions (VNFs) can be instantiated over the distributed infrastructure of one ore more network operators. Such operators need to provide high performance connectivity to the VNF instances, isolating parallel network services running on top of multiple network substrates from each other. A promising tool that can provide a dynamic connectivity framework for such a challenging environment is RINA, the Recursive InterNetwork Architecture. RINA can be used as a network virtualization solution to support the connectivity needs of VNFs hosted within the network operators' infrastructure; while bringing capabilities that current network architectures need to add as ad-hoc solutions, such as i) support for mobility; ii) multi-homing and iii) flexible mapping of application QoS requests to internal network policies. However, all existing implementations of the RINA architecture to date target Linux Operating Systems supporting generic applications; which have very different requirements than high-performance servers running VNFs on operator DCs. On this regard, this paper presents the design, implementation and initial benchmark of a software-based, performing RINA implementation specifically designed for Hypervisor systems running Virtual Machines with demanding performance requirements.

Document Type

Preliminary Edition

Language

English

CDU Subject

621.3 Electrical engineering

Subject

Xarxes d'àrea extensa (Ordinadors); Software Networks; 5G/6G & Internet of Things; RINA; Network Functions Virtualisation

Pages

6 p.

Publisher

IEEE

Version of

EUCNC 2022

Documents

PrePrint_RINA-based-virtual-networking-solution-for.pdf

598.0Kb

 

Rights

© 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

This item appears in the following Collection(s)