The Magic of Multi-cloud Networking

As I was thinking about what to write for this article, the surprising finding that multi-cloud networking is one of the top three most exciting technologies this year, according to our latest research, came to mind. I wondered why, so I started digging to understand the technology from the bottom up.

I’ve spent days exploring how it works and what’s really going on ‘under the hood.' I’ve stared at diagrams, listened to podcasts, and read white papers and documentation. Don't worry; this won’t become a lecture on MPLS and secure tunnels or split-proxy models and anycast because you don’t really need to know any of that.

Which is kinda the point.

I needed to know because, well, I have this weird block that I can’t write about a technology unless I really understand it. And by understand, I mean that I can trace packets and know which protocols and technologies are being used - and where - to make it all work.

But you don’t, and in fact, you shouldn’t need to. Multi-cloud networking is supposed to answer the challenge of “multi-cloud is just too darn complex” with something that’s easy. Ridiculously easy, even. If you need to know all the implementation details to make it work, then multi-cloud networking isn’t solving the problem at all.

That problem is the complexity of tools and APIs and networking across multi-cloud environments. In fact, that complexity is the top challenge this year, causing frustration with multi-cloud across industries, geolocations, and roles within IT.

You see, hybrid IT is the way it’s going to be for between 60-70% of enterprises out there. That means they’re operating IT stacks that are on-premises but include multiple stacks in public clouds and, increasingly, at the edge. There are all sorts of reasons for that. There’s technical debt, systems that can’t move because data has gravity, and skill deficits that prevent greater expansion across public clouds. But the reasons aren’t really all that relevant because they all point to the same inevitable conclusion: IT is, and will be for the foreseeable future, hybrid.

And that means there’s a need to seamlessly stitch together the core layers of the IT stack. That’s infrastructure, apps and app delivery, and the data that fuels the business engine. Those layers have to operate across core, cloud, and edge, and they have to do so in a secure way. They have to respect data sovereignty and the continuing hybrid app portfolio that still includes both traditional, old-school app architectures as well as modern, microservices-based apps.

Multi-cloud networking is the answer to those needs. It creates a secure mesh across properties that enables control while reducing complexity by simplifying connectivity at the network and application layers. Because it isn’t just about being able to stitch locations together, it’s about being able to control which locations have access to apps and data in all those locations.

The magic of multi-cloud networking is in the implementation details, but it manifests as multi-cloud middleware that simplifies operating across multiple environments and locations.  

And I guess that explains why folks tell us that multi-cloud networking is exciting. After all, who wouldn’t be excited about a little magic that makes managing multi-cloud a lot less complex?

Related articles: