What is MPLS: What you need to know about multi-protocol label switching

The thing about MPLS is that it is a technique, not a service, so it can offer anything from IP VPN to Metro Ethernet. It is expensive, so with the advent of SD-WAN, companies are trying to figure out how to optimize its use against less expensive connections such as the Internet.

Have you ever ordered something online from a distant retailer and then tracked the package as it makes strange and seemingly illogical stops across the country?

That is similar to the way IP routing works on the internet. When an Internet router receives an IP packet, that packet does not carry information beyond a destination IP address. There are no instructions on how that package should get to its destination or how it should be dealt with on the way.

Each router must make a separate forwarding decision for each packet based solely on the packet’s network layer header. So every time a packet reaches a router, the router has to “think” where to send the packet next. The router does this by referring to complex routing tables.

The process is repeated at each hop along the route until the packet finally reaches its destination. All those hops and all those individual routing decisions result in poor performance for urgent applications like video conferencing or Voice over IP (VoIP).

What is MPLS?

Multiprotocol Label Switching (MPLS) is proven and true networking technology has powered enterprise networks for more than two decades. Unlike other network protocols that route traffic based on source and destination address, MPLS routes traffic based on predetermined “labels”. Companies have used MPLS to connect remote branches that require access to data or applications that reside in the organization’s data center or company headquarters.

How MPLS works

With MPLS, the first time a packet enters the network, it is assigned to a specific Forwarding Service Class (CoS), also known as the Forwarding Equivalence Class (FEC), which is indicated by adding a short bit sequence ( the label) to the package. These classes are usually indicative of the type of traffic they carry. For example, a company could tag classes for real-time (voice and video), mission critical (CRM, vertical application), and best effort (Internet, email), and each application would be placed in one of these classes. The fastest, low-latency path would be reserved for real-time applications such as voice and video, ensuring quality is high. It is impossible to separate traffic based on performance with other routing protocols.

The key architectural point of all of this is that tags provide a way to attach additional information to each packet beyond what routers previously had.

Is it MPLS Layer 2 or Layer 3?

There has been a lot of confusion as to whether MPLS is a Layer 2 or a Layer 3 service. But MPLS does not fit neatly into OSI’s seven-layer hierarchy and is sometimes classified as Layer 2.5. In fact, one of the key benefits of MPLS is that it separates the forwarding mechanisms from the underlying data link service. In other words, MPLS can be used to create forwarding tables for any underlying protocol.

Specifically, MPLS routers establish a label switching route (LSP), a predetermined route for routing traffic on an MPLS network, based on the criteria of the FEC. Only after an LSP has been established can MPLS forwarding occur. LSPs are unidirectional, which means that the return traffic is sent through a different LSP.

When an end user sends traffic to the MPLS network, an incoming MPLS router at the edge of the network adds an MPLS tag. The MPLS tag consists of four subparts:

The label: The label contains all the information from the MPLS routers to determine where the packet should be forwarded.

Experimental: The experimental bits are used for Quality of Service (QoS) to set the priority that the tagged packet should have.

Bottom of the stack: The Bottom-of-Stack tells MPLS routers if they are the last leg of the journey and there are no more labels to worry about. This generally means that the router is an exit router.

Time to live: This identifies how many hops the packet can make before it is dropped.

Pros and cons of MPLS

The benefits of MPLS are scalability, performance, better bandwidth utilization, reduced network congestion, and a better end-user experience.

MPLS itself does not provide encryption, but it is a virtual private network and as such is separate from the public Internet. Therefore, MPLS is considered a secure mode of transport. And it is not vulnerable to denial of service attacks, which could affect IP-only networks.

On the downside, MPLS was designed for organizations that have multiple remote branches that are geographically dispersed across the country or world where most of the traffic was on the network to enterprise data centers. Today, businesses have shifted much of their traffic to and from cloud providers, making MPLS suboptimal.

Once companies transition to the cloud, the MPLS-based radio and hub model becomes inefficient because it routes traffic through company headquarters (hubs), which act as central choke points. It is more efficient to send traffic directly to the cloud. In addition, the use of cloud services, video and mobile applications has increased bandwidth requirements and MPLS services are difficult to scale on demand.

MPLS was a great innovation, for its time, but there are newer technologies that better address today’s network architectures. Software Defined WANs (SD-WANs) are designed with cloud connectivity in mind, which is why so many companies have been replacing or augmenting their MPLS networks with them.

MPLS is dead?

This question makes sense given the strong momentum behind SD-WANs. While MPLS is not dead, its role has certainly changed. Small and midsize businesses are likely to phase out MPLS and switch to just a broadband WAN, as many of them have moved to an all-cloud IT model.

Larger companies are likely to take a hybrid approach in which they maintain MPLS for legacy applications running on the network and then offload Internet traffic, such as the cloud, to the SD-WAN. MPLS will continue to play a role in connecting specific point-to-point locations, such as large regional offices, retail facilities with point-of-sale systems, regional manufacturing facilities, and multiple data centers. MPLS is required for real-time applications such as telepresence, although video-as-a-service providers use the Internet, and MPLS is not required.

Enterprise WAN architects must make a risk / reward calculation between the reliable but expensive performance of MPLS and the cheaper but less reliable performance of the Internet. Which brings us to how MPLS and SD-WAN will coexist.


Many network professionals see MPLS and SD-WAN as a proposition, but the reality is that both have a role to play in a modern WAN. One day SD-WANs could eradicate MPLS, but that’s decades away. Businesses already have hybrid computing, storage, and applications, so it makes sense for networks to be hybrid too.

SD-WAN is the application of software-defined networking (SDN) concepts to the WAN. This means the implementation of SD-WAN edge devices that enforce rules and policies to send traffic on the best route.

SD-WAN is a transport independent overlay that can route any type of traffic, including MPLS. The advantage of SD-WAN is that an enterprise WAN traffic architect can sit in one central point and easily enforce policies across all WAN devices.

In contrast, with MPLS, the default routes must be thoroughly provisioned and once the fixed circuits are up, making changes is not point and click.

But once an MPLS network is implemented, it offers guaranteed performance for real-time traffic. SD-WAN can route traffic by the most efficient route, but once those IP packets reach the open Internet, there are no guarantees of performance.

The most sensible strategy going forward will be to download as much MPLS traffic as possible to the public Internet, but continue to use MPLS for urgent applications that require guaranteed delivery. No one wants to be caught in the spotlight when the CEO’s monthly video conference with branch employees ends mid-sentence.

More about MPLS:

P telephone communications?

Join Network World communities at Facebook and LinkedIn to comment on the most important topics.

Copyright © 2021 IDG Communications, Inc.

Leave a Comment