The world is changing at a rapid pace. Not only are smartphones and technology a way of life, but there are connected IoT devices all around us, enabling everything from smart home appliances to cameras to industrial robots inside of factories.
Meanwhile, IoT devices are generating massive amounts of data (in some cases petabytes) that need to be processed and stored. So, why does the edge come into play here instead of the cloud? Edge computing provides the high-powered processing power of the cloud, but closer to the IoT device, enabling a faster turnaround time for analytics and data insights.
So, is the IoT here to stay? According to CB Insights, “the global IoT market is expected to exceed $1.7T by 2019, more than tripling its size from the $46B seen in 2013.” Additionally, “the total installed base of IoT devices is projected to amount to 75.44 billion worldwide by 2025, a five-fold increase in ten years.”
Equally impressive is the adoption of edge computing. “The global market for edge computing is expected to grow to $6.7 billion by 2022, up from about $1.5 billion in 2017, according to research firm Markets and Markets. This is just the beginning of the growth curve for edge computing and the IoT.
Since the future is undoubtedly going to include a rapid rise in the deployment of IoT devices, we’ll examine multiple applications of edge computing to demonstrate its value over cloud computing as it relates to the IoT.
What is Edge Computing?
Edge computing is a decentralized form of computing that enables connected devices to process data closer to where its created, otherwise known as the edge. Data processing can take place either within the device itself or close to the device.
Edge computing provides fast processing speeds, in contrast to the cloud, due to its close proximity to the data-generating device. In addition, applications may be expedited when processing occurs closer to where the data is collected, especially for manufacturers or in cases when IoT sensors are plentiful and highly distributed.
Is edge computing the future of computing when compared to the cloud? Let’s dive into why both will have a place in supporting the IoT.
Edge Computing vs. Cloud Computing
In the cloud computing model, data is stored and processed outside of a company’s physical network and instead in a co-located data center. When data is sent to the cloud, it can take minutes or more to process that data, which makes it less than ideal for being able to quickly act on data generated by IoT devices.
However, when edge computing is used, IoT devices can process data locally versus sending it to a distant data center. And while some may think that the cloud will be going away, that is not likely to happen. Instead, the cloud and edge computing will work in tandem to support your IoT devices. The cloud will be used to store large troves of data and edge computing will support the near real-time processing speeds that support IoT devices.
Edge Computing Use Cases
To help you better understand how important edge computing is to the IoT, let’s examine a few use cases.
Autonomous vehicles, in a nutshell, are high-powered computers that house multiple IoT sensors that collect data. Autonomous vehicles can’t react in real-time if the vehicle is waiting for a response from the cloud. It’s simply too slow for the vehicle to properly react in time—something that can’t wait. After all, lives depend on it. Cue edge computing for the IoT. By placing computing power close to the devices that generate vehicle data (on the edge), lag time is significantly reduced due to faster speeds. This gives autonomous vehicles the near real-time reaction time needed to respond to traffic signals and environmental conditions in order to keep passengers safe.
Healthcare is already realizing the benefits of the edge when combined with the IoT. Consider the wearable fitness trackers, smartwatches and glucose monitors that consumers use to monitor their own health in real-time via the cloud. However, the real benefits come into play inside the hospital, where there may be a dozen connected devices present in a patient’s room.
Just consider the massive amounts of sensitive, private patient data generated daily. Instead of sending that data to the cloud where it could potentially be accessed, it could happen on the edge, where the data could also be accessed in real-time. Fast data processing also supports real-time patient monitoring, which can not only improve patient health but save lives.
Specifically, manufacturing benefits greatly from edge computing. Just consider the factory floor, where there may be dozens, if not hundreds, of IoT sensors that enable workflows and provide real-time feedback on things like whether machinery is properly functioning or if equipment needs maintenance. The cloud’s latency would delay machine feedback that is critical to maintaining a fully functioning factory floor. With the edge and IoT, factories can get the real-time feedback they need versus when machine data is sent to the cloud.
Given the remote location of farms, many of which have limited access to bandwidth, agriculture is a prime candidate to benefit from edge computing. In contrast to full-time satellites or microwave connections, edge computing provides a stable, cost-effective alternative. Smart farms can use edge computing and the IoT to track equipment locations, environmental temperatures and equipment performance.
The future of the supermarket will include “smart shelves” with the help of the IoT. By applying RFID (Radio Frequency ID) sensors to shelves, retail companies are able to gather data. In some cases, robots scan the shelves to look for items that need to be restocked. Some retail stores are even allowing guests to walk out of the store with their items and charge their cards via a mobile application. This reduces wait time in line to check out and creates a seamless shopping experience for guests.
For oil and gas and energy companies, edge computing can enhance production and improve processes and security. Without plentiful connectivity in many remote locations, energy companies struggle to process data fast enough to meet their needs. Consider sensors that monitor oil pumps in the field that need to be closely monitored with human oversight today. Real-time data processing would enable energy companies to be alerted to problems with these oil pumps in order to avoid unnecessary monitoring overhead costs and improve issue response time.
From vehicles to robots, the use of IoT devices is growing rapidly. While the cloud has been the traditional method for data storage and processing, it has its limitations. As more things are connected to each other, data generation will only grow, especially as 5G is rolled out over the next five years.