In a previous post, we discussed cloud computing and how it is an effective ally when commissioning any IoT project.
Over the past few years, there has been a rise in the use of edge computing devices. While these edge devices still leverage the cloud for communicating with the outside world, they have a greater emphasis on handing primary computing tasks closer to the place where it is actually needed.
Edge computing has roots that go back to the early days of modern computing and presents a solid compromise between centralized and decentralized computing models. With a dependence on the cloud to relay data to external controllers and systems, the main computing power is happening local to the source of the data being processed.
So what is the difference between edge computing and cloud computing?
To better understand what edge computing means, think of your network as a circle where critical data is gathered and processed on the perimeter of the circle sending only “need-to-know” updates for central processing. On the contrary, in a cloud computing model, data is gathered on the edge and centralized for processing and redistribution. As an example of cloud computing, think about when you take a breath through your nose (the edge) and the air travels to your lungs (cloud) for processing. On the other hand, a true edge model would have the inhaled air being directly distributed to the bloodstream and just sending a signal to the lungs saying, “Hey, I took a breath.”
The further data has to travel, the greater the risk is of loss and latency- two things that are intolerable in a building controls system where seconds matter. According to IoT For All, edge computing is more important than ever for IoT applications. There are roughly 10 billion IoT devices worldwide, and that number is expected to grow to 22 billion by 2025- so scalability is key. Consider your home WiFi network. The more devices you have running and transmitting data concurrently, you more likely you are to experience issues with speed and connectivity. Deploying IoT with edge computing capabilities reduces your susceptibility to those same issues by handling the computing and controls tasks locally to the very same device it is installed at.
In the case of lighting, you might traditionally have an occupancy sensor that detects when you walk into a room. The occupancy sensor transmits a signal to a centralized controller that processes the data, understands the room is now occupied and tells the lighting in that area to turn on. Modern, edge-based controls manage and reconcile all that data at the sensor and local fixtures with the only data being transmitted being the fact that the lights are now, in fact, on.
As with any technology, edge computing is not without its drawbacks. Faster operation, increased efficiency and lower costs are key benefits, but each computing device you have on site does present itself as an entry-point to your local network that could be vulnerable to malware and ransomware attacks. Therefore, it is crucial that you consult with an expert who can guide you through your edge computing system design to make sure your entire network is commissioned in the most secure fashion.
If you are in need of a partner to help advise on your edge computing strategy, get in touch with us today and we will be happy to help.
In a previous blog post, we covered the basics of commercial Electric Vehicle (EV) chargers. With multiple options to choose from, there is a deeper discussion to be had around which charger type is the best fit for your facility and business type. Things like average dwell time and local driving demographics will certainly factor in your decision-making process, but the single most important aspect for all prospective buyers is cost.Read Story
As we usher in a new year (and decade), it is standard to project what’s ahead and discuss new trends in technology that will continue to shape the world we live in. Our reliance on technology and Internet connectivity as a whole has outpaced anything anyone might have imagined when Internet-connected computing entered the mainstream a few decades ago.Read Story