When you are reading about “the Edge,” do you always know what is meant by that term? The edge is a very abstract term. For example, to a service provider, the edge could mean computing devices close to a cell tower. For a manufacturing company, the edge could mean an IoT gateway with sensors located on their shop floor.
There is a need to categorize the edge more, and fortunately there are some approaches. One of them is coming from the Linux Foundation (LF Edge), which is categorizing the edge into a User Edge and Service Provider Edge. The Service Provider Edge means to provide server-based compute for the global fixed or mobile networking infrastructure. It is usually consumed as a service provided by a communications service provider (CSP).
The User Edge, on the other hand, is more complex. The computing devices are highly diverse, the environment can be different for each node, the hardware and software resources are limited, and the computing assets are usually owned and operated by the user. For example, in the sub-category On-Prem Data Center Edge, computing devices are very powerful and can be rack or blade servers located in local data centers. In the Constraint Device Edge sub-category, microcontroller-based computing devices (which you can find in modern refrigerators or smart light bulbs) are being used.
Overview of the Edge Continuum (based on the Linux Foundation Whitepaper
The Smart Device Edge
Currently, one of the most interesting Edge tiers is the Smart Device Edge. In this edge tier, cost-effective compact compute devices are being used especially for latency critical applications. For me, this is the true Internet of Things edge computing tier. You find these devices distributed in the field, for example in remote and rugged environments as well as embedded in vehicles.
Why is it hot right now? For me there are 3 key factors.
- New Use-Cases are emerging because the technology is ready for them. Or new technology was built to enable new use-cases. In either way, uses-cases such as automated theft protection, predictive maintenance, autonomous driving, digital signage and remote expert assistant can now be implemented using the latest technology developments.
- New Technology. The smart edge devices can be equipped nowadays with powerful computing hardware for a fair price and a suitable form-factor. Compare what your smartphone could do 10 years ago from now! The most important aspect here is that a graphics processing unit (GPU) is already that small and powerful and can be embedded in the device. This enables applications to leverage AI/ML, computer vision and AR/VR.
- Extending Cloud-native applications. The hardware of devices in the Smart Device Edge is not as powerful as a server in the data center, however, it is capable of containerization and virtualization, and therefore supports cloud-native software development which will play a large role in the future. Workloads and new features can be extended from the Cloud to these devices with edge native applications.
Edge Native Applications
Edge native applications are defined as “an application built natively to leverage edge computing capabilities, which would be impractical or undesirable to operate in a centralized data center” (see definition). These applications are usually distributed on multiple locations and therefore must be highly modular. They also need to be portable in order to run on different types of hardware in the field, and need to be developed for devices with limited hardware resources. These applications are increasingly leveraging cloud-native principles such containerization, microservice-based architecture, and Continuous Integration / Continuous Delivery (CI/CD) practices.
Another challenge is the deployment and management of these applications. A suitable application lifecycle management is important to provide horizontal scalability, ease the deployment, and even accelerate the development of edge native applications.
Learn More at Cisco Live
If you want to find out more, do not miss this Cisco Live session below! How do I operate applications across 1000s of locations? How do I run AI apps, on a set of nodes with limited CPU and memory? These questions will be answered!
- Edge-Native: The New Paradigm for Operating and Developing Edge Apps, BRKNWT-2205:
Frank Brockners, Distinguished Engineer, discusses questions such as, how do I operate Apps across 1000s of locations? And, how do I run AI apps, on a set of nodes with limited CPU and memory? This session will provide an understanding how to design and operate Edge apps.
- The Road to Net-Zero: How emerging technologies accelerate cracking the use-case of the century, BRKETI-1001:
Organizations need tools to steer their net-zero trajectories and address regulations, climate risks and operation resiliency. Guillaume Sauvage De Saint Marc, Sr Director Engineering, Emerging Technologies & Incubation, discusses how AI, edge, data, and cloud native technologies will sense, process and control data for emission scope, and will connect people and applications with new workflows.
Sign up for the DevNet Zone Cisco Live Email News and be the first to know about special sessions and surprises whether you are attending in person or will engage with us online.
Find Related Resources
- Try the IOT & Edge Computing learning track. Learn how to manage and deploy IoT devices and edge applications.
- Visit the DevNet IoT Dev Center to find developer resources
- Learn more about Cisco Edge Intelligence and how to control the edge with Cisco IOx
We’d love to hear what you think. Ask a question or leave a comment below.
And stay connected with Cisco DevNet on social!