MPLS Network Architecture

Recently, we have published different articles that involve explanations of MPLS and their comparison with some other networking technologies. However, we have not actually looked deep into the MPLS Network Architecture.

MPLS is a packet forwarding algorithm that relies on labels, or layer 2 information, to make its packet forwarding decision. This Label contains information about the destination of the traffic and this label helps in forward the traffic to the intended host.

Traditional IP packet forwarding algorithm primarily rely on layer 3 information to determine the path for forwarding packets; however, MPLS using layer 2 information has improved the speed and security of this process.

In this blog post, We will delve deep into MPLS network architecture, specifically its implementation in a wide area Network (WAN).

We will not cover much of the terminology and technology associated with MPLS; rather, we will focus on its implementation. We already have a post that explains MPLS in detail.

MPLS Network Architecture of a WAN

To better grasp the real-world application of MPLS, let’s examine a practical implementation in a Wide Area Network (WAN) setting.

Our WAN comprises three branch offices, each connected to a router that serves as its gateway to the internet.

MPLs network architecture

In this implementation, we introduce the concept of the MPLS Cloud, which symbolizes a series of interconnected routers within the WAN. This MPLS Cloud is responsible for handling packet forwarding within the Wide area network.

When traffic originates from any branch office, the gateway router (Ingress router) is an LER (Label Edge Router) and is responsible for the ingress and egress of labeled packets into and out of the MPLS network. It encapsulates the packet into a label. The labeled packet is then pushed into the MPLS cloud, where it efficiently traverses the network using label-based routing.

The MPLS cloud consists of a series of interconnected routers serving as label switching routers (LSRs), which is referred to as an MPLS domain. The part by which a traffic followed to get to its destination in an MPLS network is known as label switched path (LSP) and this LSP is established based on the labels assigned to packets at the ingress router

ALSO READ:  Everything About Carrier Ethernet

As the packet approaches its destination, the last router before the host decapsulates the label, exposing the original layer 3 information.

This final layer 3 data is used to route the packet to the intended destination.


With what we have covered so far, you have seen what MPLS architecture looks like. The one we covered in this post specialized in wide-area networks. There are many ways in which MPLS finds application; it can be used in a LAN or in the Internet service provider’s network. However, once you have a core understanding of what we covered in this post, understanding others will be a no-brainer for you.

If you want to learn more about some of the concepts we mentioned in this article, you can read our detailed post on Multiprotocol label switching.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top