By Austin Bankston
Edge computing is transforming the way data is being handled, processed, and delivered from millions of devices around the world. Gartner defines edge computing as “a part of a distributed computing topology in which information processing is located close to the edge – where things and people produce or consume that information” (Gartner, 2020). Due to the exponential growth of Internet of Things (IoT), edge computing was created to connect to the internet for either receiving information from the cloud or delivering data back to the cloud. Instead of relying on the cloud at one of a dozen data centers, edge computing is computing that is done at or near the source of the data.
According to Grand View Research, Inc., the global edge computing market is anticipated to reach $43.4 billion by 2027 (Grand View Research, 2020). The market is forecast to exhibit a compound annual growth rate of 37.4% over this seven-year period. While the market for edge computing, and more specifically for edge computing analytics, is still developing, some analysts and venture capitalists predict it is “the next big thing.” Edge computing is an exciting development in network infrastructure that is only just beginning to reach its potential. While it’s easy to find explanations about what edge computing is and how it works, most managers want to know how it could benefit their businesses. Several industries such as healthcare, financial sector, and the automotive industry stand to benefit immensely from edge computing.
The biggest benefit of edge computing is the ability to process and store data faster, enabling more efficient real-time applications. For example, before edge computing, a smartphone scanning a person’s face for facial recognition ran the facial recognition algorithm through a cloud-based service. A cloud-based service takes drastically longer to process, whereas an edge computing model can do so remarkably faster. Given the increasing power of smartphones, the edge computing algorithm could potentially run locally on the smartphone itself, or on an server or gateway of a network.
With this increased speed of performance, applications such as virtual and augmented reality, self-driving cars, smart cities, and even building-automation systems will be enhanced. These exciting applications of edge computing are explored in the next section.
While driverless cars are not estimated to take over the highways anytime soon, the automotive industry has already invested billions of dollars in developing the technologies to make them viable. For these vehicles to operate safely, they must be able to gather and analyze massive amounts of data. This data is pertinent to understanding the car’s direction, weather conditions, and surroundings all while communicating with other vehicles on the road (Felter, B., 2019). The driverless cars will feed data back to the manufacturers to track maintenance alerts, usage, and their interface.
Unfortunately, this influx of transmitted data will go into the same flow of traffic produced by cellular phones, personal computers, and a range of other connected devices. Due to an influx of vehicles gathering and transmitting data, bandwidth strains are inevitable if manufacturers do not adopt new computing solutions. Although an inconvenience, when an office computer experiences lag when accessing a network, it causes little risk. On the other hand, when a self-driving car experiences lag while traveling at 65 mph on an open highway, it poses a significant risk. Edge computing architecture prevents this risk by making it possible for autonomous vehicles to collect, process, and share data between vehicles and networks in real time with almost no latency.
Another application of edge computing is found in the healthcare industry. One of the biggest struggles the healthcare industry faces is integrating the latest information technology (IT) solutions with their legacy systems. Edge computing offers innovative opportunities for optimizing patient care. With the vast amounts of patient-generated health data coming from IoT devices, doctors and nurses can potentially gain access to critical information about their patients immediately, rather than waiting for results from slow and incomplete databases. Additionally, applying edge computing in this setting will allow for the improvements and evolution of medical devices themselves. This technology could be applied in creating equipment that gathers and processes data throughout the course of a diagnosis or treatment (Chen, M. et al, 2018). In addition, these applications could have considerable influence on the delivery of healthcare services in hard to reach rural areas.
Another application of edge computing can be found in finance and banking. For example, rapid data movement supports User and Entity Behavior Analytics applications that track what users are doing and how data is moving, flagging and interrupting any unusual transactions. (Technologent, 2019). For high-volume finance firms dealing in hedge funds and other markets, even a millisecond of lag in a trading algorithm computation can mean a substantial loss of money. The movie “The Hummingbird Project” demonstrated how even a millisecond lead on a competitor is worth billions of dollars. Edge computing architecture allows businesses to place servers in data centers near stock exchanges around the world. By doing this, they are able to run resource-intensive algorithms as close to the source of data as possible. In turn, they have the quickest access to data, improved response time and the most accurate information to stay ahead of their competitors.
These examples of applications of edge computing provide a glimpse at where this technology will take us in the future. Despite these intriguing applications, edge computing is still in its preliminary stage and the each of these applications rely on data science, as discussed in the next section.
Edge computing in data science
Because edge computing accelerates the data stream, it allows data scientists to process data in real-time with decreased latency. Analytics performed at the edge can identify which data to move and which to store for further analysis in the future. High-value data can also be compressed, further reducing the overall data volume and network bandwidth that’s needed to move data to the cloud.
Additionally, edge computing has created an opportunity for enterprises turn massive amounts of machine-based data into actionable insights. The IoT analytics being executed at the edge are becoming progressively more complex, with a change away from rule-based actions and more toward artificial intelligence and machine learning. By analyzing data that companies are gathering from their IoT devices near the source, data scientists can perform tasks like inventory analysis, benchmarking, predictive analytics, machine learning, and eventually, artificial intelligence from any location. Overall, it is exciting to see a glimpse at the potential applications for edge computing in the field of data analytics.
Benefits and drawbacks
For businesses seeking competitive advantage, edge computing holds tremendous possibilities. Unlike consumer applications in the home, where latency or time lag on performance is relatively unnoticeable, in some business applications this latency can become detrimental to operations. In addition, edge computing reduces IT costs because high-frequency IoT data don’t need to be moved to the cloud nor kept on-premises for analysis or long-term storage. The benefits of speed are also observed in self-driving cars. A self-driving car cannot afford the slightest delay, as its operation consists of making split-second decisions based on the surrounding environment of the road. With edge computing, network congestion will not pose problems for the cars’ feedback and operation. Overall, edge computing allows such devices to compute and make decisions locally, at high speed, without affecting their efficiency.
Edge computing can reduce latency, as there is less distance and fewer network points-of-presence between the edge and the actual location where much of the underlying data is processed and analyzed. (Moore, R., 2020) For example, the time it takes to send a query through a device to the network and for a reply to come back is called latency. Smart devices like Alexa, Google Home, or Ring have latency; they cannot operate at the desired speed of human thought. These devices are constrained by elements such as network speed, bandwidth, and even the distance of the device from the server or the database. The physical advantages from the proximity of edge devices improves real-time data analytics and lowers barriers-of-entry for on-premise hardware used in real-time applications like Virtual Reality.
Edge computing also enables real-time data analysis, providing the benefit of on-the-spot analysis. (cite) As an example, imagine being a manager of a manufacturing plant making vehicles. The manager could greatly benefit from analyzing the plant’s data as it’s being recorded, rather than having to wait for the data to go to a central server to be analyzed and then sent back to the plant. Speed of the analysis translates to immediate action, thus leading to cost reduction and/or increased revenue, key objectives for most businesses (NexGenT, T., 2020). Edge computing offers the ability to identify issues practically immediately, saving a significant amount of time and resources!
Another benefit of edge computing is the security of its data storage. The security in edge computing is the most important part of the data storage application because data is not traveling over the network; it stays where it is created. Data breaches are on the news every week! Even the most secure databases are not entirely safe. This is because data is often stored in cloud services that are breachable, allowing hackers access to users’ information, as in recorded Zoom meetings whose security has been compromised (Harwell, 2020). With edge computing, devices collect a substantial amount of information; yet, only pertinent and processed data is sent to the cloud. In several applications of edge computing, devices are not permanently connected to the network. This means if a data breach occurred in the cloud, the hacker would not have access to all of a user’s data. Therefore, with less data being sent to the cloud, there is a smaller possibility that the information can be intercepted. Although edge computing is not entirely free of security risks, it lowers the amount of data that might be at risk in the cloud.
In essence, edge computing offers a quick and reliable service that is not limited by network or bandwidth issues. When it comes to edge computing, localized information offers key benefits. Local devices or users create data that then gets uploaded to the main data center that allows for easy analysis and faster results.
Although its benefits are vast, edge computing still has drawbacks. The first drawback of edge computing is security breaches. Edge computing relies upon small data centers and the IoT, which can present several security concerns that other than traditional cybersecurity approaches can’t address. A completely secure network is not feasible, and will take many years to form. However, by adding advanced software and hardware into these devices and also empowering them analyze their data locally, all of these devices become more vulnerable to malicious attacks (NexGenT, T., 2020). Networks utilizing edge computing will have to rely more on security through the network itself.
Another drawback to the application of edge computing is possible loss of data. The technology behind edge computing requires software advancements and impressive operating power. Due to this, in the process of uploading data to the cloud, there could be a loss of user data. Considering the advancements in devices’ computational power and storage, the problem with uploading data might resolve itself in the future.
Businesses that rely heavily on data that comes from IoT devices would see the most significant benefit from adding edge computing to their business plan. Transitioning to the cloud as an IoT connected business model could be highly problematic while the data is being gathered and scaled with devices. Edge computing is designed to put applications and data closer to devices — and their users. Additionally, edge computing could help keep costs down and build processing capabilities. This technology is transforming the way data is being handled, processed, and delivered from millions of devices around the world. Edge data centers help achieve faster analysis, lower latency, regulatory compliance around location, and data privacy.
Chen, M., Zhou, J., Tao, G., Yang, J., & Hu, L. (2018). Wearable affective robot. IEEE Access, 6, 64766-64776.
Felter, B. (2019, May 10). 5 Use Cases To Know for Edge Computing and Autonomous Vehicles. Retrieved April 14, 2020, from https://www.vxchnge.com/blog/edge-computing-use-cases-autonomous-vehicles
Gartner. (2020, January). Edge Computing. Retrieved April 1, 2020, from https://www.gartner.com/en/information-technology/glossary/edge-computing
Grand View Research. (2020, March). Edge Computing Market Worth $43.4 Billion By 2027: CAGR: 37.4%. Retrieved April 1, 2020, from https://www.grandviewresearch.com/press-release/global-edge-computing-market
Harwell, D. (2020, April 3). Thousands of Zoom video calls left exposed on open Web. Retrieved April 4, 2020, from https://www.washingtonpost.com/technology/2020/04/03/thousands-zoom-video-calls-left-exposed-open-web/
Moore, R. (2020, March 10). Living on the Edge: What You Should Know About Edge Computing. Retrieved April 15, 2020, from https://www.digitalrealty.com/blog/living-on-the-edge-what-you-should-know-about-edge-networking
NexGenT, T. (2020, April 17). What is Edge Computing and why you should care about it. Retrieved April 18, 2020, from https://blog.nexgent.com/edge-computing-care/
Technologent. (2019). A Look at Use Cases for Edge Computing. Retrieved April 15, 2020, from https://blog.technologent.com/use-cases-edge-computing