Edge Computing Primer

In short, Edge Computing describes the processing and storage of data as close as possible to the place where the data is produced.

What is Edge Node and Edge Continuum?

Edge Node is a device, a server or a set of servers, that are capable to process data closer to the user. Thus, one can refer to an IoT controller or connected car as an Edge Node. Therefore, it is important to understand Edge Nodes in the context of Edge Continuum. Edge Continuum is the spectrum of possible locations where edge nodes can be placed. It starts at the smallest node (device Edge) and extends all the way to the public cloud or central data center. Depending on the use case, one or several types of Edge nodes might be required (see Figure 1).

Figure 1: Edge-Cloud Continuum

What is the difference between Edge Computing and MEC (Multi-access Edge Computing)?

The term Edge Computing describes the general concept of bringing compute and storage resources closer to the customer. Multi access Edge Computing (MEC) focuses on implementing Edge computing in the context of integration with the connectivity service provider’s network. This unlocks additional possibilities for the end user, such as access to the provider’s network services via APIs, guaranteed end-to-end QoS, integration with the service provider’s access network (e.g. 5G) and many others. The definition was coined by the ETSI. 

There are many ways to deploy Edge Computing capabilities. While the technology is still maturing, new deployment scenarios may emerge over time.

Private Edge: Industries and corporations can benefit from the advantages of Edge Computing by deploying private Edge data centers at critical locations such as big plants, headquarters or remote sites.

Colocation and hosting: Another option is to use colocation and hosting services from IT infrastructure service providers, if they are close enough to the customer’s location.

Edge Cloud: This deployment scenario combines the flexibility of cloud computing, such as abstraction of virtual infrastructure from physical, on-demand provisioning and scaling of virtualized resources, with the benefits of Edge computing, such as low latency and security.

AI Inferencing: Use cases, where pre-trained AI models must be applied for real-time data analytics and fast decision-making. Both high throughput and low latency are critical for such applications.

Low Latency Requirements: Applications that demand real-time processing and responses, such as autonomous vehicles or industrial automation.

Data Sovereignty: Scenarios where data must be processed locally to meet regulatory requirements or to enhance data security, such as in healthcare or financial services.

Scalability and Cost Efficiency: Cases where
scaling cloud infrastructure for large scale data processing would be cost-prohibitive, such as in IoT networks with thousands of devices.

Bandwidth Optimization: Situations, where transmitting large volumes of data to the cloud is impractical or too expensive, like in remote or bandwidth-limited locations.

According to a survey by UNLOCK-CEI (“Which benefits of Edge computing are most important for your projects?”), the following are the biggest benefits of Edge computing.

Figure 2: Benefits of Edge Computing

There is no one size fits all solution for the Edge architecture since different use cases and applications have very different requirements. There are several initiatives that aim to develop standardized architectures for different use cases, such as Industrial IoT or autonomous driving. Also, multiple standardization bodies have developed a series of standards for Edge Computing, like ETSI MEC standards.

Many different industries can benefit from Edge computing by moving data intensive or latency-critical workloads to the Edge. Here is a non-exhaustive list of possible use cases for Edge computing:

Agriculture

– Precision agriculture
– Livestock monitoring
– Crop health monitoring

Automotive & Transport

– Autonomous vehicles
– Connected Cars
– Traffic management

Energy

– Smart grid management
– Predictive maintenance for power infrastructure
– Distributed energy resource management

Healthcare

– Remote patient monitoring
– Real-time health data analysis
– Medical device data processing

The three main advantages of Edge computing that distinguish it from the public cloud are: low to ultra low latency, data sovereignty and lower connectivity costs. Edge Computing can be used in combination with public cloud or central data centers, that provide its users almost unlimited storage and compute resources, while Edge data centers are typically smaller and are designed for local workloads (see Figure 3).

Figure 3: Benefits of Edge versus Cloud Computing

Edge Computing is rapidly maturing, driven by advancements in 5G, IoT, and AI. However, the ecosystem is still evolving, with varying levels of maturity across providers. To choose a mature partner, prioritize companies with a proven track record in Edge deployments, robust security practices, scalable solutions, and strong industry partnerships. Look for comprehensive end-to end offerings, real-world case studies, and adherence to industry standards to ensure reliability and future proofing.