Many organizations would like to implement edge computing, but its future will be determined, at least in part, by the cost of its hardware and the development of appropriate software. At present, edge computing is very expensive to install, and only large businesses and corporations can afford these systems. Two factors restrict the use of edge computing – the expense of the hardware and the cost of creating customized edge computing software solutions.
In their 2022 Market Guide for Edge Computing, Gartner predicts that the market for edge computing will be in a state of flux for another four to six years.
Edge computing describes a range of networks and devices that monitor and work with nearby technology. The basic concept deals with processing data nearer its source, rather than sending it to a cloud that may be located far, far away. It takes time for data to move from point A to point B, and back – the greater the distance, the longer it takes (referred to as “latency”). By handling the data through the use of nearby edge computers (often specialized mini-computers), more data can be processed at greater speeds. Matt Trifirio, Vapor IO’s chief marketing officer, said in an interview:
“The whole point of edge computing is to get closer to devices, to reduce the amount of data that needs to be moved around for latency reasons, to get closer so that responses are faster.”
The use of edge computing provides results more quickly than the cloud.
Should the hardware (edge computers and sensors) used for edge computing become more expensive, as a result of economic and transportation issues, the installation of new edge computing systems will slow. If, however, the hardware becomes less costly, we can anticipate it will be embraced by medium-sized businesses and will grow at a faster pace, provided that the software is available.
The software problem is an entirely different issue. There is such a large variety of uses, and potential uses, that no standardized software has yet been developed. Each company deciding to use edge computing has had to create its own customized system. The huge variety of use cases has produced several innovative solutions that are unique and highly customized. It is generally accepted that, as patterns emerge, standardized software for edge computing will become a reality.
Edge computing must become less expensive and easier to use before it can become a mainstream, normal process.
Edge, Networks, and the Cloud
Most businesses run their edge computing systems on-premises, which is either hardwired into a network or uses Wi-Fi. As a consequence, public cloud vendors are not included in the equation. (Nor is paying for their services.) Businesses having remote devices working with a cellular network do not have to pay for streaming data (which can become expensive, especially if the data is in a video format).
Currently, businesses are using edge computing technology to process nearby data and avoid time-consuming transfers to the cloud (latency), while also avoiding expensive cloud services. There is a growing need for processing data in real time.
Internet of Things (IoT) devices were initially designed to send small amounts of data to the cloud, based on simple, specific functions and supporting special-purpose sensors (heart rate, GPS coordinates). IoT devices are, however, becoming more and more elaborate and often used by retail, via 5G, to process instant payment options, provide a good customer experience, and update inventories.
The differences between edge computing and the Internet of Things are fading rapidly, and soon there will be no significant distinction.
Examples of Where Edge Computing Is and Will Be Used
Edge computing is being used in several different industries, and for a wide variety of purposes. While the industries and uses may vary, the essential goal of minimizing latency is shared by all. Some of the more prominent uses are listed below:
Manufacturing Processes: The manufacturing industry has added millions of edge devices to collect data on equipment performance, production lines, and finished products. Edge computing provides the processing power where it’s needed. Additionally, edge devices may be programmed to transfer collected data to a central system and/or respond by initiating actions necessary at an endpoint.
Edge computing is commonly used to initiate predictive maintenance, custom production runs, and smart manufacturing. Manufacturing plants are also using edge computing to monitor and manage their energy usage.
Improving Customer Services: Businesses ranging from retail to banking are exploring the uses of edge computing with the goals of providing personalized experiences and targeted advertising. With the help of edge computing, they are also developing ways to support modern, new services, such as augmented reality shopping.
Self-Driving Cars: Autonomous vehicles (self-driving delivery trucks, self-driving cars) are an excellent example of edge computing use cases. A primary goal is their safe operation, which requires the ability to analyze all the incoming data needed for driving in real time. A real-time analysis from the cloud would simply take too long for split-second decisions. The potential for latency by sending data to the cloud for analysis means unsafe delays.
Smart Cities: City governments are using edge computing to create smart cities that use intelligent traffic controls, available parking communications, and energy usage controls. Additionally, edge computing can be used to process data from power grid sensors, public facilities, and even private buildings.
Content Delivery and Streaming Services: Edge computing is used for content delivery and streaming video because it supports low latency. It can also support good user experiences for features such as content suggestions, personalized experiences, interactive capabilities, and search functions. Some media organizations are using edge computing for the delivery of original content and live events.
Improved Security: Edge computing can be used to improve security. Businesses can use it to support video monitoring, as well as biometric scanning. Video data can be processed in real time to identify authorized individuals for entry. Organizations can use optical technologies for iris scans. Edge devices instantly analyze images of the iris to confirm matches with workers having authorized access.
On the other hand, edge computing adds devices to the system, which in turn require additional security measures.
The Promise of Edge Computing
If a standardized form of edge computing software is developed for small and medium-sized manufacturing businesses, the use of edge computing will increase significantly.
Combining machine learning with edge computing has the potential to improve the decision-making process used by edge computers. (Based on symptoms, an edge computer could learn to turn a piece of equipment “off” before it burned out.)
The number of smart cities will steadily increase. Smart cities can use a variety of sensors to improve their citizens’ lives. Pollution sensors can be used that alert city officials if an organization has exceeded its limits. Sensors for sound levels are available for some communities. Sensors can be used to analyze pedestrian and vehicle traffic, with the goal of optimizing walking and driving routes.
Zheng Song, a University of Michigan-Dearborn assistant professor in their College of Engineering and Computer Science, said in an interview:
“I believe technology-wise, edge computing and AI will be the two main opportunities for reshaping public services and infrastructure. The massive amount of sensors deployed in the field can already provide useful insights for public safety, disaster relief, smart transportation, social welfare and other domains related to public services. Edge computing along with AI will process and make sense of the huge amount of data generated by these sensors in a real-time, privacy-preserving fashion, which will unleash the true potential of such data.”
Image used under license from Shutterstock.com