What Is Edge Computing? Definition and Types

Computing was originally done using one large, centralized computer that often took up an entire room or section of a building. People would either travel from their offices to use the computer or send punch cards with programs to the system’s operator, who would input them into the computer. 

Soon, users could have their own personal computers, then personal devices, bringing a significant portion of computational processes to, or at least closer to, the edge.

In the old days, we had one big, central machine that people logged in to in order to take advantage of computational power. Users would connect to this central device and use it to perform tasks and then disconnect.

The advent of the microchip paved the way for personal computers. With personal computers, users could have all the computational power they needed sitting right on top of their desks. There was no longer a need to go to a central computer to perform essential tasks.

With cloud computing, the computational power resides in a data center away from the user—or in “the cloud.” We are currently in the cloud computing era. Much of our computing, communications, and even some of the software we use is cloud native. With an internet connection, users can interact with these resources without having to over-rely on the computational power of their own devices, which have, as a result, gotten smaller and more convenient.

Edge computing involves positioning data storage and computation closer to where it is needed. This results in improved response times and less bandwidth usage, the prime benefits of edge computing. The better response times stem from the shorter distance data has to travel while being used in processes. Instead of going all the way to a central server, as is the case with cloud computing, data can make a relatively short trip to a processor positioned at the edge, such as within a factory. 

The reduction in bandwidth that edge architectures experience is a result of less data having to travel over the internet to remote data centers. Instead, it goes back and forth between devices and computational resources closer by, which is one of the primary reasons why we need edge computing.

Edge computing is used in almost every industry now. Because faster processing time and the optimization of data flow improves nearly every organization’s infrastructure, many have adopted edge computing environments. Further, IoT devices often use edge computing for their most basic functions, which makes edge computing a compelling environment for any business that uses or sells IoT devices. 

Here are some edge computing examples.

Edge computing aids in the manufacturing process because edge devices can provide information to machines, robots, and users quickly and without using a lot of bandwidth. For example, scanners can be used to check the status of a vehicle being built as it travels along an assembly line. Users can leverage this information to improve processes and make them safer.

Edge computing plays a major role in the healthcare system because much of patient care depends on immediately available information. Edge devices are used to instantly convey data regarding the vital signs of patients, allowing doctors and nurses to make important decisions quickly and with accurate information.

The transportation industry benefits greatly from edge computing because of the proliferation of useful information that vehicles and drivers can use to increase safety and enhance the experiences of travelers and drivers. Vehicles with self-driving technology can take input from their surroundings and other vehicles and use them to make decisions. Some of the data they collect and use either comes from or gets sent to the cloud, while other data is processed at the edge.

The agricultural industry leverages edge computing to enhance the processing of data while reducing bandwidth requirements to improve how crops are grown, taken care of, and harvested. Also, data regarding the health and performance of animals, such as dairy cows, can be processed to better inform production expectations, the care of animals, and the management of energy resources that support the farm.

Telecoms have been and will likely continue to be one of the most prominent beneficiaries and providers of edge computing. Because telecommunications organizations help companies set up networks, they rely on edge computing topology to enable a wide range of devices to connect to the organization’s network and function near its edge. Everything from virtual reality headsets to gaming devices to IoT devices on manufacturing floors interact with edge computing topologies set up by telecoms. 

Further, a telecom can set up a distributed cloud that links a series of on-premises servers designed to support complex edge computing setups.

Edge computing comes with a few significant drawbacks as well. Not only can edge devices fail or lose their connection, but without the computational power of cloud-based resources, some applications simply cannot provide adequate performance within an edge topology.

If an edge device fails, there is often no redundancy in place to maintain business continuity. The end-user would have to have a backup edge device connected to the same computational and storage services. In many situations, this kind of resiliency plan would be prohibitively expensive.

Many edge devices do not have the power needed to do complex computing. A cellphone, for example, while powerful compared to what was produced decades ago, still pales in comparison to even a mid-range laptop when it comes to power. The capabilities of a data center can further dwarf the potential of the majority of edge devices.

A dependency on connectivity is an inherent flaw of all edge topologies. The infrastructure that supports many edge devices still has its foundation in cloud data centers. If the connection between the edge device or network and the cloud is lost, the topology may fail to perform altogether. 

Further, if an edge device loses its connection to the computational resources that support it, in many cases, it could be rendered useless. This holds true even in edge setups that do not rely on the internet to function.

FortiNAC enables network administrators to securely decide who is allowed access to an edge network. FortiNAC’s network access controls provide you with full visibility over your network, including each user and device that connects to it. It enables you to profile each device and user in detail. FortiNAC studies their behavior, enabling it to detect anomalous activity that could present a threat to your system.

With FortiNAC, you can also take advantage of microsegmentation policies to alter the configurations of your wireless products and switches. FortiNAC is set up to integrate with the products of over 70 vendors.

FortiNAC also gives you the ability to automate how your system responds to threats. In this way, you can prevent the east-west movement of threats once they are detected. With FortiNAC’s automation policies, you can custom-design how you respond to different types of threats, enabling you to maintain a more secure network without compromising uptime.

Also read more about Serverless Computing -- a cloud architecture that allows organizations to get on-demand access to the resources they need.

Edge computing involves positioning data storage and computation closer to where it is needed.

Edge computing is used in manufacturing, healthcare, transportation, farming, telecommunications, and other industries.

Edge computing allows you to compute with lower latency, save bandwidth, and use smart applications that implement machine learning and artificial intelligence.

The disadvantages of edge computing include the increased attack surface it creates with each edge device, a dependency on an internet-based network that could be compromised, the limited capabilities of underpowered edge devices, and a dependence on network connectivity.

Edge computing involves positioning data storage and computation closer to where it is needed.

The reduction in bandwidth that edge architectures experience is a result of less data having to travel over the internet to remote data centers. Instead, it goes back and forth between devices and computational resources closer by, which is one of the primary reasons why we need edge computing.