What is MPLS?

The Multiprotocol Label Switching ( MPLS ) protocol has been the predominant means by which service providers and enterprises built wide area networks (WANs) over the past decade. The protocol sits between layers 2 and layers 3 and was designed to support multiple types of transports.

As with IP routing, the MPLS device at the edge of a network receives incoming packets and does a route lookup. But while IP routing works hop-by-hop with the route lookup determining the next router, MPLS determines complete path to the destination. The edge device appends this path to the packet so subsequent devices can forward packets without additional routing lookups.

MPLS vs. SD-WAN: What’s the Difference?

MPLS is one type of data service that can be used by software-defined wide area networks (SD-WANs). Other types of data services that might be part of an SD-WAN include Internet access delivered across xDSL, cable or 4G. The SD-WAN is a mesh of encrypted tunnels across these underlying services, applying application routing and optimized path selection technologies to direct packets into the right tunnel. Once in that tunnel the underlying data services, such as MPLS, decides determines the path to the destination. (Visit this page for more information about SD-WAN operation.)

What are the Technical Benefits of MPLS?

Initially, the primary benefit of MPLS came about because routing lookups was difficult to implement efficiently in software. By minimizing them, MPLS offered a significant reduction in latency. Improvements in silicon have largely eliminated lookup problem, but MPLS still brings three major benefits:

MPLS traffic engineering allows fine-grained control over how the network routes traffic. The network engineer can then prevent traffic congestion, manage line capacity, and prioritize services more effectively.
Multiple service support by MPLS allows the creation of one network for all applications. Until MPLS, organizations maintained separate networks for voice and data, for example, which was very expensive.
Network uptime improves with MPLS because of its meshed design and through Fast Reroute, the ability of MPLS to recover from an error in less than 50ms.

MPLS services reflect these benefits. Generally, MPLS services are highly available (99.99% is typical for a service level agreement (SLA) and actual network availability will often be higher) and engineered to deliver low packet loss (.1% is typical of SLAs). As such, MPLS is particularly well suited for carrying real-time applications, such a voice, and thin clients that require a highly responsive network.

What are the Limitations of MPLS Services?

The several limitations when comparing MPLS services against Internet services:

Agility and costs – Enterprises are fed up with the provisioning delays of MPLS services. They look at the relatively low cost of Internet bandwidth as compared with MPLS bandwidth and the speed at which it takes to install a connection and, again, wonder why MPLS can’t be more like that.
Geographic reach – The networks of MPLS services provider do not reach into every region of the globe.
The Internet and cloud – More enterprise applications have shifted to the cloud. It makes less sense to bring traffic back to a central hub and then send the traffic onto the Internet, which is what’s commonly done with MPLS, than sending the Internet-bound traffic directly onto the Internet. Yes, there are security implications, and those need to be addressed, but the principle remains true.
Availability – Local loops are still the region most susceptible to cable cuts and outages. There’s nothing inherent in MPLS services that allow for organizations to protect themselves from these outages.

These limitations have given rise to SD-WANs. Visit this page to better understand how SD-WANs address the limitations of MPLS services.

How Does MPLS Compare with Leased Lines?

MPLS provides business with a number of concrete benefits when compared with other private data services, such as leased lines. These include:

Better performance by using Class of Service (CoS, also called QoS) and priority queuing so the network ensures the most important traffic takes priority over other traffic.
Reducing WAN operating cost s by up to 50% while maintaining a high level of reliability and service.
“Future-proofing” the WAN so it can respond to changing business needs.
Reduced packet loss means faster response for many applications, particularly real-time applications, such as voice, which are sensitive to packet loss and jitter.
Network survivability because of the fully meshed nature of MPLS backbones. Features like Fast Reroute also help.
Network consolidation providing a single, enterprise-wide view of all sites.

Technical Features of MPLS Services

In addition to the above mentioned items, the following is a list of key technical features provided by any MPLS service:

Support VoIP, real-time and bandwidth intensive Citrix applications as well as best-effort data.
Allows traffic to be “engineered” through the implementation of COS/QOS across the network and the routers. Customers should be able to manage traffic priorities.
A comprehensive, end-to-end, carrier grade service level guarantee. All equipment maintenance will also be provided with a service level agreement.
Expert advice and professional services to improve the use of technology and overall cost of the network.
Scale to allow sites to be added and bandwidth to be upgraded easily – ideal for companies changing through acquisition or consolidation.
Let your staff securely connect to your corporate network using a VPN, at the cost of a local call from anywhere in the country.
Fully meshed to flatten the topology of your network reducing the technical risk associated with a hub-and-spoke frame-relay architecture and improving overall performance.
Any-to-any configuration. When your Private Network is provisioned between all your locations, all locations can potentially connect to each other – improving the overall performance and reliability of the network.

Types of MPLS Data Services

The flexibility of MPLS has led to the development of four different types of services that may be provided by MPLS providers.

Pseudowire

Pseudowire is an industry term for transport of any packet over an MPLS network. With a pseudowire, the packet are encapsulated in MPLS and the Label Distribution Protocol (LDP) is used as a signaling mechanism. Cisco calls this Any Transport over MPLS (AToM) and this is the building block of Layer 2 VPNs over MPLS.

L3VPN

Probably the most well know MPLS service, L3VPN is an IP based VPN. Virtual routing domains (VRFs) are built on CE-router. The VRFs contain customers route information and exchange routes with PE-router, usually using a protocol like BGP.

Virtual Private Wire Service (VPWS)

With VPWS, the MPLS provider delivers the virtual equivalent of a leased line, leaving the customer responsible for managing the IP routing. If there are disparate transport types, internetworking must be done at the provider edge (PE), translating between data transport types.

Virtual Private LAN Service (VPLS)

With VPLS , the provider delivers the equivalent of a LAN to the customer. VPLS is a layer-2 offering, using MAC addresses and layer-2 switching. Customer’s sites effectively are in one large Ethernet broadcast domain with any-any-connectivity. This provides a number of benefits:

Benefit #1: Agility

The main advantage of VPLS is customer controls its own IP routing, letting the IT department be much more agile in responding to customer demands. VPLS networks allow you to rapidly reconfigure the network without having to contact and wait for your MPLS provider to act upon the request. Even if you do require a service provider change, the typical time to make network changes to Layer 2 VPLS networks is a fraction of layer-3 MPLS because network planning is much simpler, which could be crucial for some businesses.

Another feature which aids agility is the ease of adding new sites. With a VPLS-enabled network, a new site can be added by simply changing the network router that connects the site to the VPLS network. With layer-3 MPLS solutions, adding new sites is a much more complex process as all of the service provider’s routers need to be changed.

Benefit #2: More Efficiency

Companies with a VPLS-enabled wide area networks will be more smooth-running and thus should be able to provide a better level of service to their customers . This is down to the fact that with VPLS the company has access to its own network information so faults in a VPLS network can be isolated much faster and the IT department can troubleshoot to fix an urgent crisis rather than having to go through a number of support engineers to get the information required from a carrier. Less network down-time means higher corporate efficiency and productivity.

Another aspect of VPLS solutions is that they offer five levels of Quality of Service (QoS) and allow you to define your own priority levels either through labeling your traffic or using the service aware QoS feature on the core network. This is how VPLS maximizes efficient network usage according to your business needs, so you can rest assured that mission-critical data such as CRM, ERP and SCM are allocated enough bandwidth, alongside key services such as video conferencing and telephony, even during peak usage and without costly over provisioning of network capacity.

Benefit #3: Lower costs with VPLS

Companies that use VPLS solutions will find they have lower costs for a number of reasons. Firstly, VPLS enables convergence of services such as VoIP, video etc. so that all traffic can be delivered over a single Ethernet interface, eliminating multiple leased lines and resulting in economies of scale. Secondly, working with VPLS uses the same skills sets that LAN specialists have, so you would not need to provide additional training on WAN skills or hire WAN specialists. In addition, VPLS requires a lower cost CPE as it requires smaller and fewer routers than MPLS solutions.

Benefit #4: Lower latencies with VPLS

As a switched, Layer 2 solution VPLS is zero-hop in the core of the network, so extremely low round-trip latencies and jitter can be achieved. For example sub 1 millisecond within a metropolitan area and 67 milliseconds round-trip from London to New York. This improves the productivity of the workforce as information is available faster. It also saves retail customers using Point-of-Sale systems time dialling up to make credit/debit card payments, improving their customers’ sales experience.

MPLS Interconnection Points

The point at which an interconnection between two different MPLS networks is made is referred to as an MPLS Exchange Point (MEP). A straightforward MEP would be at the customer premises.

The customer must know what the COS markings mean on each side of the MEP and how to configure the interconnection router to apply them. The customer (or their consultant) also must know how to use BGP to interconnect his router with the provider-edge routers in both networks and propagate the necessary VPN routing information.
The MEP router de-encapsulates the VPN data from its MPLS headers in network A and applies the appropriate COS and MPLS encapsulation for egress into network B. While packets belonging to each COS are between networks A and B they are kept separate at the link layer within the MEP router.
If multiple MPLS VPNs are being interconnected, their data must be kept segregated from each other. There are a number of techniques for isolating this traffic, including:
Using link layer traffic isolation mechanisms, such as virtual local area networks (VLANs), frame relay, or circuit cross-connect.
Using a separate MEP router for each MPLS-VPN instance.

Interconnecting MPLS networks requires subject matter expertise. Network professionals must understand MPLS internals, vendor-specific switch/router implementations and carrier-specific COS markings. Additionally, the enterprise must be willing to build MEPs at points where carrier MPLS networks come together—typically carrier-neutral collocation facilities, or pay for local loops to the enterprise facility. Unless you have the expertise in-house, it makes sense to outsource this configuration and management.

MPLS Pricing: What to Consider

When you ask for an MPLS price quotation , you should understand that there are a wide variety of pricing models used by the different carriers. This makes it a bit more complicated to compare the carrier offerings, since with some pricing models, seemingly minor changes to your network, will change your monthly cost.

The different pricing models are as follows:

Port and local loop pricing with no charge for CoS. This gives you the most flexibility, with access to all CoS at no extra charge.
Port and local loop pricing, plus a different price for each different level of CoS, based on the bandwidth allocated for each CoS. There is nothing wrong with this approach. Just understand if you increase your bandwidth allocation for a higher priced CoS (e.g. VoIP), your monthly cost will increase.
Port and loop pricing, plus a single CoS price. This is basically option 1, plus a fixed fee for Class of Service, without regard to how you allocate the different CoS levels.
Port and local loop pricing, with “best effort” CoS included at no charge, plus additional charges for additional Classes of Service, if you need them. For instance, if you use VoIP and video, this CoS would be charged as an additional amount.

Understand that international MPLS prices may include a line item for “International Bandwidth” which reserves your contracted amount of bandwidth “across the pond” so to speak. By the same token, other carriers will bundle this charge amongst the other charges in their quotation.

As you can see, it takes some thought to analyze your MPLS price quotations.

Additional Information

For additional information about MPLS, check out the following: