The edge isn’t some place; it’s every place

[AlisonGolan]

The edge is everywhere we are — everywhere we live, everywhere we work, everywhere human activity takes place.

A few years ago, I pictured the edge as a faraway place – out in nature somewhere. Well, things have changed. Today, the edge isn’t some distant place; it’s every place.

A recent article in CIO.com, Edge Computing is Thriving in the Cloud Era, written by HPE experts Al Madden and Denis Vilfort describes this change. The article gives a brief history of the edge and then details how today’s edge technology is bolstering profits and reducing risk – all while improving products, services, and customer experience.

The historical perspective of edge in this article is well worth the read, but what I found really interesting was the authors’ description of how edge computing is being used today. Below I highlight two key points the authors make about why edge computing is thriving in today’s cloud era: the importance of real-time analytics and distributed computing.

Real-time edge analytics at the edge

Compared to a few years ago, it’s amazing what businesses can do at the edge. That’s because data is now generated from copious amounts of sensors and cameras and then analyzed at the edge extremely powerful, distributed computers — all at reasonable costs.

Here’s a few examples listed in the article:

• Supermarket fraud prevention

Many supermarkets now use some sort of self-checkout, and unfortunately, they are also seeing increased fraud. A nefarious shopper can substitute a lower priced bar code for a more expensive product, thereby paying less. To detect this type of fraud, stores are now using high-powered cameras that compare product scanned and weight to what they are supposed to be. These cameras are relatively inexpensive, yet they generate a tremendous amount of data. By moving computing to the edge, the data can be analyzed instantly. This means stores can detect fraud in real time instead of after the “customer” has left the parking lot.

• Food production monitoring

Today, a manufacturing plant can be equipped with scores of cameras and sensors at each step of the manufacturing process. Real-time analysis and AI-driven inference can reveal in milliseconds, or even microseconds, if something is wrong or if the process is drifting. Maybe a camera reveals too much sugar is being added or too toppings cover an item. With cameras and real-time analysis, production lines can be tuned to stop the drift, or even stopped if repairs are required – without causing catastrophic losses.

• AI-driven edge computing for healthcare

In healthcare, infrared and X-ray cameras have been game changing because they provide high resolution and deliver images rapidly to technicians and physicians. With such high resolution, AI can now filter, assess, and diagnose abnormalities before getting to a doctor for confirmation. By deploying AI-driven edge computing, doctors save time because they don’t have to rely on sending data to the cloud to get a diagnosis. Thus, an oncologist looking to see if a patient has lung cancer can apply real-time AI filters to the picture of the patient’s lungs to get a quick and accurate diagnosis and greatly reduce the anxiety of a patient waiting to hear back.

• Autonomous vehicles powered by analytics

Autonomous vehicles are possible today because relatively inexpensive and available cameras offer 360-degree stereoscopic vision. Analytics also enable precise image recognition, so the computer can decipher the difference between a tumbleweed and the neighbor’s cat – and decide if it’s time to brake or steer around the obstacle to ensure safety. The affordability, availability, and miniaturization of high-powered GPUs and CPUs enables the real-time pattern recognition and vector planning that is the driving intelligence of autonomous vehicles. For autonomous vehicles to be successful, they must have enough data and processing power to make intelligent decisions fast enough to apply corrective action. That is now possible only with today’s edge technology.  

Benefits of distributed computing at the edge

Edge computing is a form of computing done on site or near a particular data source, in contrast to sending it to a remote data center or the cloud to be processed. As a result, edge computing lets business process data faster and more efficiently.

In the past, edge points generated massive amounts of data that often went unused. Today, extremely powerful (yet affordable) computers are deployed at the edge, allowing companies to optimize operations better. Because data is distributed across edge locations, issues are addressed in real time without having to worry about delays caused by connecting to the cloud.  

According to the article, we’ve come a long way since the first wave of edge technology.  Now more companies than ever before can harness comprehensive data analysis without the IT infrastructure needed in previous generations.

Today’s edge technology is not just aiding companies bolster profits, but in fact, it’s helping them to reduce risk and improve products, services, and the experiences of people that engage with them.

To learn more  about how data can be analyzed and used at the edge in real time, check out the website, Intelligent Edge: Edge computing solutions for data driven operations. To understand what happens at the edge, at the core, and in between, read this blog on how HPE Ezmeral Data Fabric provides a modern data infrastructure that empowers data-driven decision making at the edge.

Hewlett Packard Enterprise

twitter.com/HPE_IoT
linkedin.com/company/hewlett-packard-enterprise
hpe.com/info/iiot