Software-defined wide-area networking (SD-WAN) and multiprotocol label switching (MPLS) are both technologies on the highest level of the enterprise networking hierarchy. While the former is fairly new — rising to prominence in the late 2010s — and the latter first came on the scene about 20 years ago, each style of wide area network (WAN) can efficiently and effectively handle traffic at the scale demanded by the largest organizations.
Some in recent years have decided that SD-WAN is superior to MPLS across the board. That isn't necessarily true. This Techtorial examines these networking solutions — how they differ from each other, their advantages and disadvantages, optimal applications for both tools and more.
The first factor that differentiates SD-WAN from an MPLS network is right there in the former's name: While small, router-like devices are often used to help direct traffic and signals traveling over a virtualized SD-WAN network, its key strengths are in the software at its core. By contrast, the success of an MPLS solution is contingent in large part on the hardware and infrastructure supporting it.
Think of it this way: SD-WAN effectively functions as a virtual private network (VPN) overlay atop one or more networks. As we will see later, that one system can easily be an MPLS circuit. It may also use a combination of several different network types to facilitate an internet connection, including standard broadband, LTE, 4G or 5G cellular, ethernet and others.
An MPLS solution can stand on its own and offer efficient management of high-volume network traffic, with fairly guaranteed quality of service (QoS) and considerable privacy. SD-WAN does not function on its own in quite the same way — but it can make less expensive broadband networking much better. An advantage like that will always be attractive for many users.
The impact of MPLS, upon its debut in the late 1990s (after more than a decade of development), cannot be overstated. For one, it was the first large-scale enterprise networking solution that could offer a truly private VPN connection. But even more importantly, its use of revolutionary packet-forwarding and labeling technologies allowed it to direct network traffic with greater speed, quality and efficiency than frame relay and its other predecessors. Packet loss, once all but a given, became increasingly infrequent. (For a meticulously detailed explanation of the labeling protocols in MPLS and other aspects of its WAN architecture, see the Internet Engineering Task Force's original document on the subject.)
MPLS is arguably best known and appreciated for its reliability. Customers who choose MPLS know they can count on providers to stick to the terms of their service-level agreements (SLAs) and maintain high QoS. They also know they can expect strong packet delivery and low latency, even with rapidly fluctuating WAN traffic demands across networks that span multiple countries or continents.
Large corporate and educational campus-style networks, metropolitan-area ethernet setups and cloud providers handling a great deal of private traffic are all major examples of MPLS use cases. Most large organizations requiring a guaranteed high QoS can benefit significantly from the modern deployment of a traditional MPLS network. In the enterprise arena, which can often be unpredictable and sometimes downright chaotic, MPLS is eminently dependable.
MPLS's longevity in its field is remarkable, given how fast things move in telecom and how quickly products and methodologies can become obsolete. But if it were perfect, there would be no competition for its market share. Thus, it's important to address the needs that SD-WAN can meet more effectively than MPLS.
SD-WAN's structure as a VPN overlay that streamlines and accelerates the traffic coming through different internet circuits makes it much more flexible than an MPLS solution can be on its own — particularly in terms of scalability. For example, if a business ever needs to reduce or increase the scope of its MPLS architecture, it can be expensive and difficult to do so. Cost is one of the biggest motivating factors behind the decision of some organizations to move away from MPLS-only networking structures (and other, more traditional WAN deployments) and toward SD-WAN.
The ability of SD-WAN to juggle multiple networks also means it can route traffic appropriately through them. It does so based on which connections are functioning most effectively — or according to the needs of network administrators and end users — through dynamic path selection. This allows admins to perform "application steering:" prioritizing certain network applications (such as video conferencing) and their traffic over others, based on levels of urgency. As is the case with its scalability, SD-WAN's flexibility comes at a lower financial and resource cost than trying to accomplish the same things with standalone MPLS. (Part of this stems from MPLS' reliance on the data center. That limitation puts it somewhat behind the eight-ball in a business world that is increasingly reliant on the cloud.)
Furthermore, SD-WAN addresses the need for total control over traffic and bandwidth more effectively than other networking solutions. Admins have unfettered oversight of traffic patterns, and can amass data about these trends in real time for later use in various WAN optimization efforts. They can also respond more quickly when users experience connectivity and performance problems, which boosts customer satisfaction.
Lastly, the built-in data protection offered by SD-WAN means it can provide a level of security equivalent to — and in some cases greater than — what MPLS can manage, often doing so at a lower total cost.
This is a complicated question, and we understand why company leaders and IT professionals are asking it.
The answer is that SD-WAN may eventually replace MPLS outright. But it won't do so in the immediate future. This wholesale replacement might not necessarily even happen in the next five to 10 years.
Why not? Primarily because organizations that have used MPLS for years seem to have developed an "if it ain't broke, don't fix it" mentality. In fact, the vast majority of MPLS customers plan to either expand or significantly increase their infrastructure during the next several years — although plans can always change, especially in the tech world.
There are certain areas where SD-WAN has replaced MPLS. Retail is perhaps the most notable example, as SD-WAN's dynamic path selection is critical to handling high-volume traffic from customers' transactions. Healthcare and finance organizations are also migrating to SD-WAN in significant numbers. But even if SD-WAN were to phase out MPLS, certain technologies are already viewed in some circles as replacements for the former: Secure Access Service Edge (SASE), a cloud-based, distributed architecture networking method, is the most likely candidate to be considered "the next big thing" in enterprise telecom.
By keeping MPLS infrastructure in place and adding SD-WAN on top of it (and perhaps some lower-level broadband options as a backup if these are not already there), you have all the dependability of MPLS while also giving yourself greater network control. Using this hybrid WAN setup to steer traffic as dynamically and efficiently as possible benefits you and your customers alike.
GTT Communications has the expertise, breadth of services and ingenuity to help you choose the best enterprise telecom solution for your networking needs. As a global managed service provider tapped into Tier 1 infrastructure, we're capable of deploying and overseeing SD-WAN overlaid on broadband internet, SD-WAN with MPLS, MPLS alone or other WAN connectivity types. To learn more, get in touch with us!
Transform your WAN with dynamic network traffic managementLearn More
Leverage anti-virus, firewall and anti-malware services all in one scalable solutionLearn More
Defend against cyberthreats and block access to malicious sitesLearn More
Monitor your cloud environment, enforce policies and mitigate shadow ITLearn More
Strengthen remote access and prevent unauthorized intrusion with identity- and permissions-based authentication policiesLearn More