Scalability testing of legacy MPLS-based Virtual Private Networks

MultiProtocol Label Switching (MPLS) have been the basic building block for Layer 2 and Layer 3 Virtual Private Networks (VPNs) for almost 20 years. While MPLS data plane is quite simple, the control plane has become increasingly complex, and therefore costly. The typical Label Switch Router (LSR),...

Full description

Saved in:
Bibliographic Details
Published in2017 IEEE URUCON pp. 1 - 4
Main Authors Lopez, Gustavo, Grampin, Eduardo
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2017
Subjects
Emulation
Legacy VPNs
Memory management
MPLS
Routing
Routing protocols
Scalability
Software
Virtual private networks
Online AccessGet full text

Cover

More Information
Summary:MultiProtocol Label Switching (MPLS) have been the basic building block for Layer 2 and Layer 3 Virtual Private Networks (VPNs) for almost 20 years. While MPLS data plane is quite simple, the control plane has become increasingly complex, and therefore costly. The typical Label Switch Router (LSR), implements both forwarding and control plane in the box, which comprises distributed routing protocols such as OSPF or IS-IS for routing information gathering and dissemination, distributed signaling and label distribution mechanisms such as RSVP-TE and LDP, and also MP-BGP (Multi Protocol BGP) for actual VPN routing implementation, as defined by well established IETF standards. The stress imposed on backbone routers by the diversity and scale of services is one of the reasons for the emergence of the Software Defined Networking (SDN) paradigm, seeking to liberate routing hardware from the burden of service implementation, transferring the control plane complexity to logically-centralized controller software. Given the interest in replacing legacy MPLS-based VPN (and other) services, quantifying the effect of such replacement seems logical for service providers planning the migration. A first step is to actually measure the scalability of legacy implementations; to this end, we built an emulated environment using a commercial implementation and performed a series of testing, which results are presented in this paper.
DOI:10.1109/URUCON.2017.8171874