What is Intelligent Edge Computing?

You’ve heard of edge computing. You may even have set up an edge architecture using 5G or a platform like Kubernetes. But do you have a mere edge architecture or an intelligent edge architecture? That will be one of the questions organizations ask themselves going forward, as the idea of intelligent edge computing increasingly comes to the fore.

Here’s a primer on what intelligent edge computing means and why it’s becoming important today.

Defining intelligent edge computing

In a nutshell, intelligent edge computing is the application of edge computing architectures to workloads involving data analysis, machine learning, or AI.

Generically speaking, an edge architecture is one that places data or applications at the network edge, where they can be accessed by users with fewer latency or reliability delays caused by the network.

Edge architectures can be used for any type of workload. For instance, a retailer could use edge infrastructure to process purchase transactions in local stores, which would prevent purchase delays or disruptions caused by problems reaching centralized servers in the cloud. Or, a business could store its backup data in a data center located in the same city, where the data could be downloaded faster in the event that it needs to perform a recovery.

These two examples are edge computing use cases. But they aren’t forms of intelligent edge computing because the workloads they entail don’t involve smart data processing or analytics.

Examples of intelligent edge computing

To deploy an intelligent edge computing workload, you need an application that analyzes data in some way.

One example is network-connected cameras that monitor a home, then use facial recognition AI to figure out who’s in the home. If they detect the presence of unknown individuals without the homeowners on-site, they could flag it as a security event.

In this case, the ability to process the data right at the edge, rather than having to move it into the cloud, process it there, and send back results, would enable faster decision-making — a potentially critical factor in a security-sensitive use case like this.

A car that uses sensors to analyze the physical environment is another example of an intelligent edge use case. Cars can generate 25 gigabytes of data in a single hour. If they had to move all of that data into the cloud and back in order to apply AI to it, the results might be meaningless for vehicles that need to make split-second decisions based on the data.

Related: Container Technology Comes of Age at the Intelligent Edge

Is it really different from edge?

You could argue that the concept of the intelligent edge is really just a buzzword that doesn’t add much value to existing understandings of edge computing. That would be a fair assessment, to an extent. It is a category of edge computing, or one set of potential use cases for edge architectures, more than it is a fundamental paradigm unto itself.

Still, in a world where data has little value unless you can process it automatically and quickly, it’s easy to see how the intelligent edge is on track to become the predominant form of edge computing writ large. There will always be other edge use cases, but perhaps nowhere are edge architectures more valuable than in situations where you have a lot of data that needs to be analyzed quickly, without waiting on the network to move it.

For that reason, expect to hear more and more about intelligent edge computing as businesses leverage edge environments to process data more efficiently.