The term “edge” has multiple (often contradictory) definitions, which can make it difficult to fully understand. Since most organizations are already operating an edge of some sort, it is vital that we not only understand the edge, but that we also devise a strategy to enable it.
NaaS has emerged as a key enabler to make the edge more accessible, and therefore more valuable, to businesses. Let’s take a closer look.
What is the Edge?
We probably won’t ever agree on a singular definition of “the edge.” But we can at least survey some of the most prominent definitions to create a clearer picture for all of us.
In the world of computing, some of the earliest applications of the label “edge” had to do with networking. And specifically those points where two networks meet. Edge routers, or gateways, have long been the connection between two networks, living at the edge of each. But that’s not really what most folks mean when they talk about the edge today.
Real estate folks talk about “edge data centers” that are, for some, anything that is not in New York City or Los Angeles. For others, it’s anything outside of an “NFL city,” and for a few it has to be a shipping container in a field to be a true “edge DC.” Meanwhile, engineers focused on compute and storage are talking about “edge compute” that is, for some, running containers (or VMs, gasp) on a UCPE (Universal Customer Premise Equipment) type device; for others, a hardened device or set of devices to run in mines or on factory floors, etc; and for another set, it is anything that is still on premises. And now, we’ve even started talking about mobile edge, device edge, etc.—where the ultimate CPE is the edge, like your phone, your car, or your doorbell.
For me, one of the most convincing definitions is that of “edge interconnection,” which is essentially describing any location that doesn’t yet have an IXP (internet exchange point), also called a carrier hotel or a neutral meet-me-room (MMR). This encompasses much of the rural broadband challenges that have been discussed, and the solutions that have been funded, in the US as of late.
Why You Need an Edge Connectivity Strategy
Okay, so at least there is consensus on one thing: the edge is NOT the cloud. In some ways, it’s everything else. The reductionist could say that we’ve basically split our IT estates into two; the cloud and the edge. Even if you take a more nuanced or specific view of what the edge really is—or where it is—we can all agree that it is. Meaning that, however you choose to define it, the edge is a real and valuable construct that helps us define a set of challenges many of us are facing. Let’s take some examples to clarify.
A great place to start is with a CDN (Content Distribution Network) because they are, by definition, an edge service. The whole point of a CDN is to take content closer to the edge of the network so that users distant from the content source have quick access to a local cache. So, maybe that example is too obvious to be useful after all; of course a CDN needs an edge strategy. But what about a cloud provider? Why would they need an edge strategy? After all, in many ways the cloud is the logical opposite of the edge, right? Well, it turns out that even cloud providers need to provide end-to-end services to their clients. And that means, you guessed it, they must find a way to connect to their users out there on the edge. We see this in the drive behind direct connections, among other trends.
As you may guess, there are tons of additional use cases for edge connectivity between these two extremes. Also intuitive is the fact that a lot of these scenarios revolve around data and storage. Many, if not most organizations are now generating and collecting data on a simply massive scale. In many cases, this data is either captured or needed at an edge location.
Whether it’s video data for virtual green screens in video production, telemetry data from manufacturing equipment, or user data being fed through AI algorithms to provide better recommendations in real time, we need to be able to move this data from edge to cloud and back again. And we must do it reliably, often within strict latency requirements. In other words, we need an edge connectivity strategy.
How NaaS Enables the Edge
Getting to the edge is hard. Traditionally, we’ve had two options. Option one was to rely on incumbent carriers and a patchwork of transit and/or transport providers to deliver connectivity to your edge. Option two was to directly take on the burden of owning and operating the network you needed to provide the required access across your IT estate.
Both were very often slow—with service activation measured in days, weeks, or often months—and inflexible. Once a circuit was finally up, it had to stay that way; typically for years. And additional capacity required that you started the whole process over again. None of this is conducive to the demands of today’s edge, no matter how we define it.
That’s exactly where NaaS comes in. NaaS provides business agility through automation, a true as-a-service operational model, and massive amounts of resilient network capacity available on demand. NaaS enables the edge by providing dynamic accessibility to your edge, across your edge, and between your edge, your cloud, and anywhere else it’s needed.
What’s more, PacketFabric has embedded object storage into their NaaS, providing yet another edge enabler. So, you can still use NaaS to turn network capacity up and down where and when you need additional access to an edge location; and you can also store and access petabytes of data right in the same network—much closer to your edge than the cloud can reach.
The Bottom Line
While we may not ever agree on a singular definition of the edge, we can probably all agree that access to our edge is a critical business enabler. And it turns out that NaaS is the perfect solution for providing that access.