A Guide to the Main Types of Edge Computing: Exploring Variants

All types of edge computing adhere to the same basic principle. They aim to process as much data as possible at the edge of a network. As edge computing develops, however, different types of edge computing are implementing this principle in different ways. Here is a quick guide to what you need to know.

The background to edge computing

Before looking more closely at the different types of edge computing, it is worth taking a step back and looking at the background of edge computing.

In the early days of IT, there were no networks. This means that data had to be processed locally. As networks developed and became clouds, data processing became centralized. Over time, however, centralized data processing has become too much of a good thing. It is generating a lot of traffic and not really getting maximum value from data centers.

Edge computing aims to address this by processing data locally as much as possible. The key point to note is that edge computing does not (currently) aim to replace the cloud. It aims to reduce its workload. In fact, many types of edge computing make use of the cloud.

Types of edge computing

Here is a quick guide to the main types of edge computing and their practical implications.

Standard edge computing

This is generally what people mean when they just say edge computing. In the most basic form of edge computing, edge devices are totally responsible for data processing. This approach to edge computing has very clear limitations. At the same time, it can be perfectly adequate for simple tasks.

Mobile edge computing

This is essentially a variation of standard edge computing. The key difference is that it uses standard mobile devices (e.g. phones and tablets) to perform the data processing. Mobile edge computing is useful in situations where you need speed but not permanence.

For example, say a business takes on extra warehouse space at peak times. It may need to process data from the warehouse at a very high speed. It may not be able to risk any delays or issues with internet performance. On the other hand, it only uses the warehouse space at peak times. It therefore probably doesn’t want to spend money on a permanent solution.

Mobile edge computing could be a perfect choice here. It uses standard devices that most businesses already have and turns them into edge devices. The one potential downside to mobile edge computing is that you may need to invest in higher-specced mobile devices. These may be somewhat more expensive. The trade-off, however, can be more than worth it.

Fog edge computing

Fog edge computing takes the data-processing load off end devices but does not send it to the cloud. Instead, the data is processed by local devices such as gateways, routers, and switches. Fog edge computing is still relatively simple and economical to implement. It is also quick because data is still processed locally.

The main use case for fog edge computing is in environments where large amounts of data need to be processed but the processing is fairly simple. Right now, the main use of fog edge computing is in healthcare and industrial automation. It is also gaining use in smart cities.

Multiaccess edge computing

Multiaccess edge computing is very similar to fog edge computing. The key difference is that multiaccess edge computing uses proper servers instead of network devices. This makes it a more powerful option. It is generally used for applications that require significant amounts of data processed very quickly. Common examples of this are online gaming, streaming video, and virtual reality.

Cloud edge computing

Cloud edge computing is a step up from multiaccess edge computing. It doesn’t just use servers, it uses local clouds. This means that it delivers the maximum level of processing power available outside of centralized data centers.

Cloud edge computing also has the highest implementation costs. It, therefore, tends to be used for applications where huge amounts of data need to be processed with minimal delay. Use cases for cloud edge computing include the industrial internet of things, remote video monitoring, and some autonomous vehicles.

Hybrid edge computing

Hybrid edge computing is essentially the edge computing equivalent of the hybrid cloud. It combines different types of edge computing into one edge computing system. Hybrid edge computing aims to deliver maximum benefit for minimum resources. Like the hybrid cloud, however, it is so complicated, it’s really only likely to be suitable for enterprises.

Edge computing and the cloud

Edge computing can also be used with the regular cloud. The edge system would perform as much local data processing as it could. The regular cloud would then take over.