EdgeOps Fundamentals: How Edge Computing Helps You

In my previous post, I defined EdgeOps as “An Adaptation of DevOps for the Edge.” I then spent some time discussing how DevOps and EdgeOps, while close, have a few significant differences. Properly delineating those two concepts does not explain why edge computing — and, by extension, EdgeOps — is useful or relevant to you. In this post, I will introduce you to the value EdgeOps can provide by describing the IT challenges edge computing can help you overcome.

The first challenge is bandwidth usage optimization. The growing availability of 5G and other narrowband connectivity technologies — such as Dash7, LoRaWAN, and Zigbee — simplify the deployment of IoT devices wherever they are needed cost-effectively. A single IoT device can generate gigabytes or even terabytes of data every day, after all. Transmitting all that data does not make sense from a cost or business perspective.

Edge computing is also a great way to reduce latency at make it more predictable. Most corporate and industrial networks feature highly variable latency. Transmitting data straight to the Cloud over the public Internet adds another level of unpredictable latency. Naturally, deterministic local area networks provide a way to keep latency under control in a specific location, but Internet service providers cannot offer the same guarantees. Low latency kept within tightly enforced bounds is vital to mission-critical and real-time applications. Suppose you are automating a nuclear power plant, a welding robot, or a self-driving vehicle. In that case, you cannot afford to have time-sensitive decisions made in the far-away Cloud due to the latency induced by the Internet.

Governments, organizations, and individuals have recently expressed a growing interest in data sovereignty. This correlates with the many well-publicized security breaches involving IoT devices in the same period. Data handling is controlled by several layers of laws and regulations in multiple jurisdictions. You need to take into account not only where the data is collected but also the places where it is stored and even processed. Data sovereignty describes the efforts of governments to protect the data of the people and corporations under their purview. Highly regulated industries, such as healthcare, possess stringent encryption and data residency requirements for that reason. Edge computing provides infrastructure that streamlines the fulfillment of such requirements.

Finally, edge computing is widely regarded as an effective way to make applications more resilient. Deploying networking infrastructure, storage, and compute outside the Cloud and corporate data centers is an opportunity to make applications stronger when facing disruptions in their operational environment. Naturally, the level of resiliency achieved will depend on many factors, such as whether the infrastructure features redundancy or the size of the geographical area where the infrastructure is deployed.

Another benefit of Edge Computing is that it allows implementing reporting by exception. The value of an individual data point varies depending on the use case. Sometimes, it is essential to process and archive every reading in a historian database. Sometimes, the only readings that matter are those outside normal thresholds. This means that only exceptions are reported. Of course, it is possible to implement reporting by exception directly on devices driving sensors. However, it is smarter to offload such decisions to a nearby edge node if the application involves complex data models, machine learning, or video analytics. Reporting by exception is especially useful for battery-operated devices since it reduces the need to transmit data over the network, which always consumes a significant amount of power.

As you can see, edge computing can help you overcome a variety of challenges. As for EdgeOps, it simply provides an easier way to reap the benefits of Edge Computing. You see, building and deploying resilient applications that provide data sovereignty and guarantee low latency is no easy task. Deploying your application on an Edge Computing platform built from the group up for EdgeOps is one way to streamline the process.

In my next post, I will discuss the wider edge computing ecosystem and where EdgeOps fits in it. If you wish to learn more about EdgeOps, please look at our full white paper.

All Edge of Things contributors belong to Eclipse Foundation member organizations or are Foundation staff.