Layer 2 Vpn Architectures [Electronic resources] نسخه متنی

Summary

Extending LAN services across multiple metropolitan areas is one of the critical requirements for large Metro Ethernet deployment. Service providers want to leverage their existing Layer 3 WAN infrastructure to bridge the distance gap of LAN services.

VPLS is a multipoint Layer 2 VPN architecture designed to fulfill such a requirement with considerations to scalability and performance. Comparing point-to-point Layer 2 VPN architectures, VPLS is relatively more complex in both implementation and deployment.

This chapter started with an overview of VPLS architecture that defined the service requirements, explained the concept and components of a virtual switch, and identified the characteristics of the VPLS control plane and data plane.

When it comes to topologic models, VPLS is quite flexible depending on the size and complexity of individual deployment. The basic VPLS models include full-mesh and hub-and-spoke, and the more advanced hierarchical VPLS is a hybrid of both basic models. Hierarchical VPLS can have either MPLS access networks or QinQ networks. Redundancy is always a top priority for service providers. You can accomplish VPLS redundancy by using multihoming and STPs. In a Metro Ethernet deployment, you should limit the scope of spanning trees within individual islands for bandwidth efficiency and network stability in the backbone.

This chapter showed basic VPLS configuration step-by-step and gave a VPLS example that used different switchport modes on the interfaces connecting to customer networks. The case studies highlighted the scenarios commonly seen in VPLS deployment, such as per-VLAN MAC address limiting, QoS, Layer 2 protocol tunnel, backdoor links, and multihoming.