In the ongoing quest for practical “multi” solutions (multi-orbit, -mission, -satellite, -operator, etc.), I raised in my last post that the current focus on antennas was a “necessary but not sufficient” component. Achieving practical interoperations that meet efficiency and SWAP demands (not to mention economics) will also require digitally transforming ground systems/space networks with virtualized, standards-compliant modems and other components.

Virtualizing is just step one, however. Enabling fast, reliable, automated and resilient switching between those multi-variables (most of them governed by mission demands and SLAs) requires intelligent coordination. That’s the harder part and where SD-WAN may be the answer.

A software-defined wide area network (SD-WAN) is a virtual WAN architecture that allows enterprises to leverage multiple transport services. Centralized control functions steer traffic securely and intelligently based on the needs of each specific application across the WAN including partners, cloud operators and as-a-service (aaS) providers. That’s largely because different applications have different transport requirements. For example, a serious gamer cares first and foremost about latency, while someone trying to download a movie cares mainly about speed, not latency. Recognizing these individual needs to orchestrate the network is what SD-WANs are designed to do.

SD-WAN may be what ultimately integrates satellite into the larger telco world, where satellite will be one slice of a large, coordinated service delivery pie.

None of this mattered to satellite operators before clouds and virtual telecom networks, back when satellites were all GEO and broadcast ruled the market, but it’s critical in an IP transport, multi-multi-multi ecosystem. Alternatives create opportunities.

I don’t want to sound Pollyanna, however. As with most things satellite, unique complexities arise. For example, unlike terrestrial networks, satellite links can have very different round trip times creating challenges in delivering packets in the correct order. Similarly, because available data rates in LEO are variable and not necessarily guaranteed, some form of probing mechanism is needed. There are others, but there are also solutions.

So, is SD-WAN for satellite necessary? Possibly. Sufficient? Nope. Here? Close. Smart? Yes.