EVO I

The first evolution of SD-WAN (Software-defined Wide Area Network) was to enable customer to migrate from MPLS (Multiprotocol Label Switching) to commodity internet. In Australia, we have different viable carriage options such as business grade nbn (National Broadband Network) and LTE (Long Term Evolution). MPLS has been around for more than 25 years however the cost for leased line is still substantially high. The main reason for this, MPLS comes with Service-level Agreements while internet is best effort only. Fortunately, true SD-WAN vendors have technology to resolve this.

During this evolution, many Service Providers realise they were going to lose revenue from their MPLS network and started a race to the bottom. Thus, countering the cost saving justifications for moving toward SD-WAN. Some Service Provider had the foresight to see the value of productising SD-WAN. This allowed them to provide their customers carriage diversity at a great price point while lowering their own internal cost to deliver these services.

EVO II

The second evolution of SD-WAN was to help customers connect to multiple clouds through the concept of Cloud On-ramp. Many customers are currently cloud first or cloud only. They have begun decentralising their applications and have strongly adopted SaaS (Software as a Service). Therefore, traffic from the branch was no longer taking the most optimal path via the headend. This hair-pinning of traffic add extra latency and unnecessary load on expensive on-premise infrastructure.

EVO III

The third evolution of SD-WAN began with the announcement of SASE (Secure Access Service Edge). During COVID-19, there was a understanding that employees weren’t just physically working at the branch. If they were sitting in the branch, they were not there just between 9-5. In addition, most companies were forced to undergo digital transformation. Now, there is greater trust and desire to have flexible work locations via leveraging online collaboration tools.

Since applications are now decentralised and users are becoming disparate. There is a need for a new framework to protect all these users, applications, and data. Furthermore, there is a realisation that there will never be a GOD box that will be able to effectively secure the branch and forward packets at the same time. This is where SASE will truly shine.

Final Thoughts

Seeing the amazing progression in SD-WAN, I am eager to see the next evolution. I predict that it will be geared towards a combination of edge computing, 5G and self-healing networks. It is going to be exciting times for customer that has already adopted SD-WAN or should be re-evaluating that transition.