Jump to section

What is MPLS?

Copy URL

Multiprotocol label switching (MPLS) is a technology designed to enhance the speed and efficiency of data forwarding across large networks and/or at edge locations. MPLS works in a virtual private network (VPN) and integrates with any underlying infrastructure—including internet protocol (IP), ethernet, frame relay and asynchronous transfer mode (ATM), making it a scalable, low latency networking option. 

MPLS establishes a multipoint system of nodes—or routers—that carry a packaged unit of data (also known as an IP packet) from one IP to another on the most efficient path possible. A packet begins its journey on the edge of the MPLS network at an ingress Label Edge Router (LER), where the packet is examined and assigned a forward equivalence class (FEC). Based on the FEC, a label is pushed onto the packet to determine where it will travel next in the MPLS network along a one-directional label-switched path (LSP). An MPLS label header is made up of four parts.

  • The label (20 bits) contains the information about where the packet should go next. 
  • Experimental (3 bits) performs quality of service (QoS) and determines a packet’s priority level.
  • Bottom-of-stack (1 bit) lets an MPLS router know if it is the final router or egress router on an LSP. 
  • Time-to-live (TTL) (8 bits) determines how many swaps a packet can make.

From there, the packet travels to multiple Label Switch Routers (LSRs). Each time, a new label is swapped onto the packet based on the network’s lookup or routing table, or label information base (LIB). The process continues until the packet arrives at an egress LER. The LER pops, or removes, the label and forwards the packet to the new destination IP address. Sometimes multiple labels are grouped together, creating a label stack. 

Think of it as luggage transferring from one plane to another on a connecting flight. The baggage handlers stick on a label with its destination, and at each leg of the trip, the label is updated to inform where it should head next, until finally, it reaches its endpoint and is routed to baggage claim.

Pro: Unlike classic data routing—in which a router must examine the packet and browse multiple network addresses for a match—MPLS combines the most effective aspects of routing and switching to determine a fast, efficient, and reliable path. And since MPLS exists between the data link and the network layer of the OSI 7 layer model, it will work on top of any IP’s underlying infrastructure. 

Pro: MPLS works in a VPN, making it a more secure transport option compared to IP-based networks. MPLS is not encrypted, however, and users should consider adding this layer of protection to an MPLS network.

Pro: Because FECs and LIBs respectively assign a packet a number and then push it along a preset path according to that number, the packet's entire journey is mapped out before it starts moving, and that makes MPLS connections more reliable. And since packet labels are quickly swapped, network traffic delays and high latency are often avoided. MPLS is especially useful in connecting the remote devices and real-time applications that make up the Internet of Things (IoT)

Pro: MPLS-Transport Profile (TP), released in 2008, makes additional improvements to the original MPLS system—including reversible paths, built-in maintenance tools, continuity checks, and protection paths in case of outages. While the original IP/MPLS works well in the core network (which connects multiple networks together in what’s known as a network backbone), MPLS-TP is best suited for the edge, aggregation, and access networks.

Con: MPLS costs more than standard internet service providers.

Con: MPLS was designed to run information through central hubs, such as data centers or corporate offices. However, with the proliferation of cloud services, it is now more efficient to send information straight to the cloud, especially large packets such as videos and mobile application data. 

Con: The label-switched paths that make MPLS a fast and predictable transport method also take a long time to set up, especially if you are expanding across multiple networks. This makes it hard to scale quickly with MPLS.

MPLS and the cloud are not a perfect match. With most types of traffic going directly from organizations to the cloud, MPLS routing from point to point can feel obsolete. However, there are ways to combine MPLS and cloud computing for specific benefits. 

Maybe your organization wants to transport some select information to and from its corporate office via MPLS. You can offload the rest of the data for the cloud, and MPLS can carry only what is needed to go to the office. Or you could enhance MPLS with broadband internet links via SD-WAN, a software-defined wide area network, and determine the best transport method for each packet according to network protocol and bandwidth needs.

SD-WAN is an acronym for software-defined wide area network. It typically replaces the connectivity of traditional branch routers with virtualized or appliance-based software that routes traffic to/from remote locations, using centralized policy control and management.

Similar to MPLS, SD-WAN routing policies are used to send traffic over the fastest connection that meets the traffic’s security requirements, application needs, and service-level agreements (SLAs). With today’s bandwidth-consuming, distributed, cloud-based applications, sometimes you might need a combination of both highly reliable MPLS and lower cost but best effort internet connections to maintain service levels and flexibility. SD-WAN makes it possible to get the benefits of MPLS while keeping up with growth in data traffic and meeting the needs of a digitally enabled world where employees, customers, and other stakeholders demand better experiences.

Whether working in the datacenter or at the edge, you need a fast and reliable transport network. Red Hat® Enterprise Linux® combines the benefits of routing technologies like MPLS and SD-WAN with built-in edge capabilities, continuous IT security, and a consistent operating foundation needed for modern IT and enterprise hybrid cloud deployments.

Red Hat Enterprise Linux gives you supported access to open source innovation by representing your requirements upstream and returning with stable updates. Red Hat has a longstanding practice of doing work upstream first, vetting ideas with the larger community and working together to build new features, releases, content, and more. 

Keep reading

Article

IoT and edge computing

IoT needs compute power closer to where a physical device or data source is located. Edge computing provides that local source of processing and storage for IoT.

Article

What is edge computing?

Edge computing is computing that takes place at or near the physical location of either the user or the source of the data.

Article

Edge computing for telco

Edge computing is a priority for many telco service providers as they modernize their networks and seek new sources of revenue.

More about edge

Products

A stable, proven foundation that’s versatile enough for rolling out new applications, virtualizing environments, and creating a more secure hybrid cloud.

An enterprise application platform with a unified set of tested services for bringing apps to market on your choice of infrastructure.

A portfolio of enterprise software optimized for lightweight deployment at the edge.

Resources

Brief

Edge computing in action: Space

Podcast

Command Line Heroes Season 8, Episode 8:
"Robots and vehicles"

E-Book

Gain a competitive edge with your container strategy