SD-WAN Explained: What Is SD-WAN and How Does It Work?

In a networked world, companies need to provide business-critical IT at multiple locations. Software-defined Wide Area Network, or SD-WAN for short, is an optimal solution for this. Flexibility and cost efficiency compared to conventional network infrastructure solutions make SD WAN attractive. How does it work and what advantages does it have over other network solutions?

1. SD-WAN explained

SD-WAN is an automated overlay technology that uses encrypted, application-specific dynamic tunnels to provide the best user and application experience with the network infrastructure over which it runs.

This technology enables companies to supplement or replace conventional WAN traffic services such as Multiprotocol Label Switching (MPLS, see MPLS vs. SD-WAN). It often uses lower-cost Internet access services to cost-effectively extend WAN connectivity with Quality of Service (QoS) to small sites. SD-WAN also allows organizations to decouple the user and application experience from the technology and provider of last mile network connectivity. This creates the impression of a unified network, regardless of the technologies used.

2. Why SD-WAN?

In the past, companies had to build their own private network or rely on traditional WAN services from providers. These solutions were not always flexible enough to adapt to changing requirements, such as the introduction of a hybrid cloud. These customer networks are more difficult to manage at scale and slower to deploy. Older solutions also offer less visibility and no customized application-level traffic handling.

But it’s not just the problems and shortcomings of traditional WAN models that have driven the adoption of SD-WAN. It is also the frustration with the procurement process in telecoms that has driven the adoption of SD-WAN. Customers want to separate the purchase of connectivity from the provision of networks. SD-WAN makes this possible by freeing them from the telco buying process and giving them more flexibility.

3. How does SD-WAN work?

Essentially, SD-WAN separates the control layer from the data layer. The SD-WAN controller, which houses the control plane, acts as the central brain of the solution. It provides the instructions and rules that edge devices need to forward packets effectively.

The controller also collects real-time feedback from the SD-WAN appliances about the traffic coming from the connected LANs and the performance of the connected lines. This allows the controller to direct traffic in real time to shape the applications to achieve better performance results. The data plane resides in the SD-WAN edge appliances located at each customer site. These appliances are connected to the customer’s LANs and network access lines. While the control plane is the brain, so to speak, the data plane executes the actions specified by the controller.

This architecture enables faster, more informed and more effective data transfer, resulting in a better and more consistent overall experience for the end user.

4. SD-WAN vs. MPLS

Although SD-WAN and MPLS are often seen as alternatives to each other, they are not mutually exclusive. There are some important differences between the two.

MPLS WAN solutions from carriers are operated via a pre-installed shared services private network architecture. SD-WAN, on the other hand, is a virtual, software-based solution that is independent of the underlying transport architecture. For this reason, it is often associated with the Internet, the most favorable transmission medium. This is an important difference, because with a traditional MPLS WAN you are limited by the access technologies you can use. With SD-WAN, on the other hand, you can mix and match IP access solutions from different providers, including MPLS, with SD-WAN technology bringing them together into a manageable network.

For example, SD-WAN can enable a customized level of resilience and bandwidth, resulting in a tailored, cost-effective solution. This also means that SD-WAN can include sites with mobile broadband access only if required. For example, pop-up sites or remote sites without staff can also be included.

As mentioned above, SD-WAN enables multiple transport options, including MPLS. MPLS is considered the best transport network solution because it can guarantee bandwidth and quality for a given amount of traffic. SD-WAN, on the other hand, can shape the traffic and select the best performing connection to meet the quality of service parameters. However, it cannot make the Internet faster. This is where MPLS comes into play. A high-performance, high-speed MPLS connection to critical locations can ensure that traffic makes short, reliable hops from source to destination via a manageable number of peering points.

5. SD-WAN advantages

Now that you know how SD-WAN works and why you should use it, here are some other benefits of an SD-WAN solution:

• Cost-effective network: SD-WAN gives organizations the flexibility to use a variety of transport options for resiliency and scale bandwidth as needed based on location.
• Any location can be connected: Organizations can use any provider to cost-effectively connect any desired location to the network.
• Increased site uptime: If there is more than one access line at a site, SD-WAN automatically prioritizes and reroutes applications when network issues are detected.
• WAN optimization: Application-aware routing improves performance, and targeted forward error correction eliminates traditional packet loss degradation.
• Network security over the public Internet: Encrypted tunnels enable secure use of the low-cost public Internet transport.
• Short lead times for new sites: A pre-configured SD-WAN appliance can use any internet connection to bring a site online quickly.
• Increased Quality of Experience: Applications get their own performance requirements, and automated error correction and prioritization deliver optimal experience.
• Centralized control and visibility: The centralized control plane enables end-to-end performance management via a customer portal.