IoT: Get your Edge on computing

There were times when people use to get a lot of buzz on cloud computing. Those times are well passed and a new buzz has taken over the internet. The term edge computing is primarily coined to include all menial networking techniques into one. This includes peer-to-peer networking, ad-hoc networking, cloud-based systems, and distributed networking systems. It refers to a technique where data processing happens at the edge of the network neighboring devices rather than routing data all the way to the cloud. The biggest advantage that edge computing brings to the internet of things (IoT) is the security by reducing the distance of data travel.

Many people would argue about the processing of a large data set. The simple answer is that data processing will happen at the nearby devices available in your network. In this form of computing, the nearby devices will be chosen based on the specifications and the necessary processing will be done locally or in close vicinity. It justifies the phrase that cloud is coming to you. The concept is propelled further by a well known model called the evergreen model. The big players of the software industry such as Google and Microsoft are already diving-in. Google Chrome is the best example when it comes to demonstrate the so called evergreen model. The model suggests that the software is always up-to-date and the user doesn’t have to install or update anything. Microsoft pushed the concept with its Azure Sphere. Another great example is Mozilla’s Evergreen Firefox browser.

Although edge computing technology has been there for a while but it is still to gain some popularity. Apple’s Live photos and Google’s motion photos does a similar job by taking photos and stitching them together in small animation (GIF) image locally before a cloud sync. Google took it a little further with its Google Clips camera in 2017 but days have already passed without any buzz. Google’s concept was to capture candid shots of people and stitch them locally into GIF images. People argued that the technology is kind of creepy as many would not want to get a candid shot. But some praised the concept but not the device. The picture quality is not up-to-the-mark and device itself is very costly.

The applications and the use of edge computing technology is more pronounced in autonomous vehicles or self-driving cars where latency is a big issue. Computing must happen locally or at the nearby devices or servers to avoid mishaps. In IoT world, smart manufacturing is taking over the outdated manufacturing techniques and edge computing will play a major role in solving the problems. The applications are not limited to the mentioned examples but can extend to an array of problems in today’s computing world.

There is a close similarity between fog computing (fogging) and edge computing (edging). Both are decentralized form of computing and both pushes computing power closer to the data source. The major difference arises when it comes to processing capability. In fogging, devices don’t have processing capability. In edging, processing capabilities are passed on to the device. It can be best understood in a way that fog computing is closer to the cloud whereas edge computing is more closer to the device. In essence, it can be concluded that edge computing is in fact based on fog computing principles. Both techniques can be adopted based on the data processing capabilities.

Industry is getting the resonance on edge computing. GE Digital recently introduced Predix Edge Manager that manages many edge devices, servers, and applications. You can easily manage your edge devices provided you have their license and product. Researchers and engineers are pushing their limits to achieve an unforeseeable future where everything is connected and managed. To what extent it goes, we just have to wait and watch.