What is Edge Computing?
“Edge computing is a distributed computing paradigm that brings computation and data storage closer to the sources of data in order to improve response time and save bandwidth.” Edge computing is often associated with Data Centers because initially that is what it was talking about. Bringing the data closer or storing it in the cloud.
Now it has moved on to various technologies and there are companies out there that specialize in Edge Computing. Nutanix being one of those companies has a convenient manual titled “Edge Computing for Dummies”. This would be an excellent resource if you are looking to implement edge computing into your business today.
How Does it Work?
Essentially, sensors, actuators and gates, get their own processing power enabling the connected devices to make their own decisions rather than a central or cloud processor handling those tasks.
This cuts down on errors that are caused by latency and improve the speed at which these devices perform their tasks. If we think of self-driving cars, then having an improved reaction time doesn’t just make it more efficient, it can save lives.
On the Edge
Edge computing has many use cases that in the short term we will mostly see utilized by the industrial and manufacturing sectors, but as time goes on we will see it seep into consumer hands especially when it comes to (IOT) Internet of things.
If you don’t know what the IOT is then let me take a second to explain it. The internet of things is a network of objects that have embedded sensors, software and other technologies that collect and exchange data. An example would be something like a smart dishwasher, a smart light and a smart toaster. Depending on your needs, all these things can work independently from one another or collectively. Internet of things sensors could be used to collect data about the temperature and humidity in your home and use this data to make real time corrections to the temperature or dehumidification system to keep your home just right.
Industrial Internet of Things (IIOT)
This is where computing is used for real-time monitoring, predictive maintenance, and automation in the industrial sector. By processing the data closer to the machinery and devices it enables faster response times, reduces downtime and optimizes operational efficiency. It also helps to extract massive amounts of data from sensors allowing manufacturers the opportunity to see where and why failures occur.
This can help with a number of things: predictive maintenance, quality control and inspection, real time analytics and optimization, robotics and automation, worker safety and ergonomics and supply chain optimization.
Predictive maintenance is enhanced by real-time monitoring of industrial equipment. Having the equipment processing its own data anomalies in the system can be identified immediately. This proactive information can prevent downtime or catastrophic errors that could be a massive detriment to the system.
Quality control sensors can quickly detect any defects or inconsistencies and deviations from quality standards. By having these sensors and cameras able to quickly process the data, operators are able to make corrective changes before wasting valuable resources on improperly made products.
Data Collected by various sensors can identify areas in which slow down occurs. This data being processed in real time can either allow for workers to move to a new area in real time to mitigate the overflow. It also allows for manufacturers to create new processes after reviewing the data that can prevent future problems.
Robots with their own processing capabilities can make decisions to boost productivity based on the current conditions. Reducing the need for a centralized processor makes the speed, precision and responsiveness of robots more efficient and flexible in manufacturing processes.
It can also help with optimizing your supply chain operations by collecting and analyzing data in real time at various points in the supply chain. It can help track inventory, monitor logistics, and analyze demand patterns. This allows for faster response times and creates a reduction in lead times, allowing you to get your product to your customer just in time. This also works in retail systems to track inventory or manage backups at checkout.
Smart Cities and Video Surveillance
Edge computing is probably necessary to maintain a smart city. In order to have the processing power and capability to run all of a city and facilitate its data in real time to make decisions about managing critical infrastructure, power, water, traffic management systems, smart grids and public safety networks many devices will have to be able to make decisions in real-time. Surveillance systems are used in many industries and in the public sector. To make these systems more efficient, having video data processed at the source reduces bandwidth requirements, latency and allows for analytics done at the point of contact. This can be incredibly useful for facial recognition and anomaly detection such as a fire in the theater.
Health Care
Remote patient care and telehealth systems could rise in popularity among healthcare providers and patients alike with the advent of edge computing. It would allow doctors to respond to changes in condition in real time regardless of distance. This could be beneficial to home virtual consults, but could also allow you to see a specialist while relaying data from one facility to another. This could also help with the quality of life to home bound patients or those in long term care facilities.
The Risks
Edge computing allows for more sophisticated systems with increased complexity, which could require more technical intervention or IT infrastructure. Due to the nature and connectivity of edge devices they are vulnerable to security attacks, because they can be in remote unsecured locations, such as a traffic light at 3 a.m. The cost of edge computing devices goes up because you have high value SOC’s in these devices in order for them to operate.
Data collection could be more difficult to recompile due to fragmentation of data. Where it would work well with one traffic light to compile all traffic lights could introduce gaps in data. This could introduce latency going the other way, and have slow downs due to bandwidth limitation on the devices. With that being said, these problems can be addressed and mitigated if systems are implemented. On a small scale some of these risks are non-existent because you are in close proximity and the devices are in a secure location.
The Edge is Near
This is just a small sample of ways that edge computing can allow scaling and benefit us in everyday life as well as create a safer, more efficient work environment.
Edge computing is still in it’s early stages of development and utilization. It is a powerful and useful new technology or at least new use of technology that improves the way we live and work. As more uses are developed and more people and places implement edge computing we should also see more innovations.
Tell me what you think about edge computing, would you like to see it at work or making your drive there more efficient? Would it being in your home benefit your everyday life?
Best,
Chris