What is Edge Computing?
At present, we have a problem with edge computing. They all speak different languages, and they say a lot, but they talk to each other. When I used to manage global development teams building distributed applications for Fortune 500 companies, I used to start a project with a team glossary. Why? Because when you can get 100 smart developers and architects in a room, you quickly find out that almost everyone has a different tech jargon, especially if it’s a Java girl trying to communicate with a PL/SQL guy. There is a great need to set a standard across the edge computing industry. No doubt, all of you are being bombarded with vendor messages, white papers, and panel discussions about edge and edge computing as the next big thing. In a way, that’s true. In other ways, it’s nothing to get excited about. So, let’s start unpacking edge computing and review the main goals needed to understand what edge computing has become and will become as it continues to evolve.
The location of Edge Computing
Where is the edge? That’s a great question. Simply put, the edge is anywhere but the core cloud or enterprise data centers. For the geeks: The edge—for edge computing’s sake—is the vast home to all the different kinds of distributed computing that aren’t cloud computing as we’ve known it for the past decade or so. Contrary to popular belief, the border is not defined by the boundaries of the network. As my friend Rob Tiffany, vice president of IoT strategy at Ericsson, says, the advantage of any application starts with the default location. It is the location of the endpoint device, whether it is a sensor, programmable logic controller, smartphone, or connected car.
In a physical sense, the default location could be a warehouse, a factory, or even a person moving through the city. It also happens to be the location where the data is generated and consumed. Yes, edge computing can be bi-directional. It’s not just about aggregating data from endpoint devices and sensors. The border extends from the default location to anywhere and anything except the central cloud. This extension is the domain of edge computing, in its many flavors and forms. For a particular distributed application, the range between the default location and the location where the server-side application and compute resources reside is the distributed or edge computing context.
To the edge or not to the edge?
When we work to move from the physical endpoint location to the proximity of the Edge Server location, we can be in proximity or out of proximity. The former are commonly considered on-premise locations, such as your home, a university or business building or campus, or a small cell or nearby cell phone tower serving a roaming mobile user. Proximity edge computing is becoming more important for mission-critical computing that may require highly reliable, low-latency connections between the endpoint client device and edge server resources. Think of a smart factory or a smart port. Security and privacy are becoming increasingly important location concerns. It is important to note that proximity edge computing can also be peer-to-peer. A good example would be wirelessly connected drones to participate in swarm intelligence and self-orchestration and management around coordinated tasks.
The out-of-proximity server location for edge computing can be at a telecommunications central office, a colocation site that houses a regional or municipal data center for a company. It can also be one of a growing range of physical or logical availability zones, local zones, and edge points of the presence of CDN (Content Delivery Network) providers. That can include the big cloud services providers like AWS, Microsoft Azure, and Google Cloud. Out-of-proximity edge computing resources tend to serve less critical and non-time sensitive edges computing application workloads, such as local content caching and delivery, bulk data aggregation, or model training of federated ML. While they do provide some degree of latency benefit, their primary value is to reduce the expensive backhaul costs of sending massive amounts of data between the central public clouds and the default location.
Technically speaking, these are network borders.
You have to consider the network or telecommunications view. This angle at the edge is important because, from a technical perspective, the computation (processing, memory, and storage) physically and logically resides on a network. Without the network and infrastructure hardware, there is no cloud, no edge cloud, and no edge computing. Computer science is the application of the software. In the case of edge computing, it has a distributed and increasingly disaggregated software architecture. On the other hand, computing is a hardware-oriented term that means the infrastructure resources on which applications run. Think of this in terms of edge servers that contain processors, memory, storage, and networking components that connect them to the network.