What is Edge Computing?

A traditional centralized IT structure is composed of two main elements. In the periphery, there are several endpoints whose main job is to capture data and act according to the instructions. The data from these peripheral devices is transmitted to and from the second element of the structure: a central server. This server is where practically all the heavy work of the system takes place; That is, where the data is processed, consolidated, analyzed and activated.

Edge Computing involves a distributed rather than centralized approach. Instead of having all computing activities occurring on a central server level, as many processing tasks as possible occur on the periphery, as close as possible to the data source. With more computing occurring at the edge of the network, it reduces the need for long-distance contact between peripheral devices and the central server.

How does Edge Computing work?

In many edge structures, major computational tasks occur at the IoT device level. See an industrial sensor, for example. Under a centralized structure, the sensor only captures data and sends it back to the server. With an EDGE approach, the sensing device can process this onboard data and automatically trigger specific actions if necessary, without having to interact with the server.

In addition, or as an alternative to device-level processing, the structure may also include deploying computers or edge servers. Therefore, instead of all peripheral devices transmitting data to a central location, the data is sent to computers that are physically closer (for example, on the shop floor). Specific and time-dependent calculations can be performed much closer to the point of need, saving time and cost of data transmission from long distance.

What is an example of edge computing?

An example is a video surveillance system. Increasingly, these systems incorporate relatively sophisticated video management software (VMs) with features such as motion detection and automated monitoring of the number of employees.

In a centralized structure, all video images are sent to a central server, where they are analyzed by the VMs. With the EDGE architecture, the immediate analysis can be done on the camera itself. For example, a smart camera will only start relaying images to a server if and when the movement is detected and is able to trigger the digital shooting wires themselves. This approach greatly reduces the amount of expensive bandwidth required to relay video images to the remote server.

What is the purpose of Edge Computing?

The main objective is usually to reduce latency. In sensitive IoT applications, for example, industrial processes or health procedures, every second or millisecond counts. By moving processing and computing as close as possible to the edge, it is possible to substantially reduce the gap between the emergence of a critical situation and the taking of the appropriate action.

Can Edge Computing Replace Cloud Computing?

Edge computing is best seen as a complement to cloud computing rather than a substitute for it.

To group data from multiple IoT devices, monitor efficiency across the organization, usage trends and performance, you will still need the data to be processed and analyzed centrally, probably in the cloud. Meanwhile, time-sensitive device-specific processing can occur at the edge of the network.

What are the benefits of Edge Computing?

  • Responsiveness:

Great physical distances, network congestion and transmission interruptions can delay the time it takes a system to respond to what critical situations can be. With Edge Computing, you don't need to send data to a remote analytics center for decisions to be made so that your IoT devices can respond faster.

  • cost:

When data is processed at the device level or relayed to edge servers, you can substantially reduce the volume of data that needs to be transmitted to distant centers. This means lower bandwidth requirements, less network congestion, as well as lower data transmission costs. This also means additional EDGE computing resources, either on the device or on an Edge computer, which results in additional costs, so the system architecture is important.

  • reliability:

In some environments (eg rural agricultural sites and offshore facilities), connectivity can be restricted or unreliable. If IoT devices can perform core tasks autonomously, it is much less likely that you will suffer business disruptions as a result of connectivity issues.

  • Privacy:

The transmission of personal data across national or regional borders for processing can lead to regulatory problems (GDPR, for example). With an edge computing approach, you can process raw data in the periphery and then obscur and protect it before passing it to data centers in other jurisdictions.

  • security:

Edge computing can help ensure that a larger slice of raw data remains at the source. Generally, there is less data being transmitted and all the data that needs to be sent can be simplified and protected more easily. This means a smaller attack surface and less chance of data breach.

What are the disadvantages of edge computing?

  • Hardware enhancements:

If more computing is done on the periphery of the network, your EDGE devices need to be properly equipped for work. This could mean significant expense in device updates or replacements. Depending on the model of the structure, there is also the cost of additional edge servers to consider.

  • connectivity:

Edge computing significantly reduces the need for long-distance data transmission but does not completely eliminate it. You still need to think of an appropriate channel for at least some minimum level of communication with the central server.

  • security:

Increasing the features of your Edge IoT devices can also lead to additional exploration opportunities for attackers. A robust approach to device management is essential, including patch distribution and appropriate encryption for data in transmission.

Does Edge Computing Need 5G?

It is probably more accurate to say that 5G allows for the emergence of some hybrid edge computing architectures. For example, distributed multiple-access edge computing or DMEC (Distributed-MEC).

With the fifth generation of cellular technology, it becomes possible to transfer large amounts of data in real time. This opens the doors to a much wider range of IoT applications, including those using artificial intelligence (AI), machine learning, multimedia streaming, and even virtual reality.

Under the 5G standard, latency (ie the delay between sending and receiving information) is reduced from 200 milliseconds from 4G to 1 millisecond. In fact, to meet this benchmark and provide the type of ultra-responsive, resource-rich services that companies and their customers expect, it is almost certain that more than cellular network updates will be needed. If large volumes of data are still being transmitted to a central server, companies are at risk of falling behind. Changing most of the processing closer to Edge will help release bandwidth and ensure you get all the benefits of 5G.

With Distributed-MEC (or DMEC), edge computing can be performed on the mobile main network. DMEC leverages the 5G ultra-low latency network and also removes latency and delay associated with Internet connections. It can also remove hardware costs on both devices and eliminate the need for any EDGE computer equipment.

 

For more explanations about IoT, visit our glossary.