The Internet is a vast decentralized network that facilitates communication between machines on a monumental scale. In fact, by 2025 it is estimated that we will produce 463 exabytes of data every day globally. That is equivalent to 5.36 quadrillion bytes per second! On such a grand scale, you may wonder how this data reaches the correct locations in acceptable times?
Well, this very question has given rise to the idea of ‘edge computing’. In simple terms, edge computing means that computer workloads are kept at the ‘edge’ of their networks. Crucially, they are kept as close to the data’s point of origin as possible. This means that data does not have to be beamed to and from a centralized server for processing. Rather, distributed computational power stores and processes the data in localized sites.
The main advantage of edge computing is that it distributes responsibility and workloads (similar to the MapReduce programming model).
Feeling a little overwhelmed with that technical word jumble? Don’t worry – this article will cover how edge computing works, the pros & cons associated with the tech, and some real-world examples.
How Does Edge Computing Work?
The defining feature of edge computing is that processing power is kept close to the edge of a network. You can think of this in terms of the physical location of devices. It makes more sense for many computers over a wide geographic area to work together, rather than one computer in a centralized location doing all the work.
Take for example smartphones. Almost everyone has one with a powerful processor capable of handling video calls, running applications, and sorting data. If smartphones did not have powerful processors, they would have to offshore their workloads to other locations. That means running computations on external servers as cloud applications.
Cloud computing is certainly favorable for some workloads e.g. cloud-based PBX. However, edge computing remains advantageous in other contexts. We’ll explain why in the next section.
Benefits of Edge Computing
Are you wondering why edge computing is so valuable? Well, this section explains the top 3 most important advantages of the technology:
Higher Bandwidth
Bandwidth can be defined as the maximum rate of data that may be transferred from point A to point B. It follows that a network with high bandwidth is able to transfer large quantities of data in both directions.
These days, workloads and devices are producing so much data that it can be difficult for the underlying network to keep up. Often, systems that produce a lot of data will become congested, resulting in long upload and download times.
To avoid network congestion, it’s actually better to store and process that data in the place where it was produced. In edge computing, that may be in an ‘edge node’ or in the device itself.
Lower Latency
Another bottleneck on networks is the issue of latency. By this, we mean the speed at which data reaches point A from point B. The Internet is fast, but it is not instant. In fact, it takes about 232 milliseconds for data to travel from London to Singapore.
For systems that rely on speed, it is better to expedite the process by cutting out unnecessary data transfers. That means keeping only pinging data to geographically proximate nodes, rather than sending it halfway around the world.
Autonomy
Edge computing means that devices are able to handle tasks even when disconnected from their network. This is useful for devices in physically remote locations that may not have a reliable Internet connection.
That means that their workflow can go on uninterrupted by the events of the outside world. Then, once the connection is re-established, the ‘results’ of the computed data may be sent to a central server for analytical purposes and feedback.
To sum it up, autonomy is guaranteed through the distribution of workload responsibility.
Challenges of Edge Computing
Whilst there are compelling arguments for edge computing, questions remain about the proper implementation of the tech:
Scalability
For edge computing to work as intended, the workload must be distributed to a wide pool of devices. That means that if you want to scale up your operation, you must also deploy new hardware to run it. Sadly, the rollout and maintenance of such hardware can be an expensive and time-consuming operation.
Connectivity
As edge computing relies on the distribution of devices over a wide geographic area, it also increases the risk that one part will be struck by an outage. Maintaining far-flung devices on the network can be challenging particularly in extreme environments.
Security
Data centers are entrusted with a large amount of sensitive data, and thus the standards for security are held very high. However, on distributed networks, the same level of security cannot be guaranteed. Some devices may be vulnerable to cyberattacks, which can be difficult for an IT team to spot and remedy.
Examples of Real-Life Use Cases
Edge-based computing is already widespread, and you most likely already interact with it on a daily basis. Here are just a few potential applications of the technology:
Workplace Software
Edge computing is often used in the handling of workplace data. For example, call center workforce management software may need to make split-second decisions about priorities for staff members. As such, an edge node should direct the operation, as opposed to a centralized server hundreds of kilometers away.
Gaming Consoles
Spend any time in the gaming community and you will likely hear complaints about ‘lag’. While the root cause of this can be attributed to several factors, one common cause is network latency.
Generally, latency should be kept to a minimum on gaming devices to provide the end user with a smooth experience. As such, gaming consoles will usually have beefy graphics processors to handle the bulk of the computations.
Digital Assistants
Voice-controlled digital assistants like Amazon’s Alexa are becoming more sophisticated (and more popular) every year. However, many people still see these as ‘background’ devices that are only there to help when needed. For this reason, companies like Amazon want to keep their network impact minimal and reduce the amount of bandwidth they consume.
Conclusion
This article has given a basic introduction to what edge computing is and how it works. By now, you should have a good understanding of the applications of the tech, as well as its bottlenecks and limitations.
Still struggling to get your head around edge computing? You might consider using immersive learning tech to delve deeper into complex topics such as this. In fact, many virtual and blended teams are embracing the tech to onboard their staff members who learn better from visual stimuli.
