As digital transformation is added to more and more company leader’s strategic plans, they are thinking about goals and how technology can be leveraged to make improvements to new or existing products and services. Over time more and more “things” are being connected, and the Internet of Things (IoT) is now often referred to as the “Internet of Everything.” More specifically, cloud computing is widely adopted across industries, and is still popular in many applications; but recently, there has been a shift toward edge computing for IoT applications. Edge computing is what happens right at the edge of the network where IoT connects the physical world to the cloud, or the world of computation. There have been a few specific edge device trends that our team has seen growth and impact businesses in various industries:
Low power connectivity
Edge devices that need to live outside a local wireless network or mesh network commonly use cellular communication to allow them to achieve connectivity. Until recently, this was the only feasible approach that provided sufficient global coverage. Today, low power, low bandwidth radio technologies like SigFox and LoRa are growing in global coverage, which is giving edge devices other feasible options for connectivity. For edge devices that don’t need high data rates, these are even better options than cellular since the progression of cellular is moving to higher power and higher bandwidth radio technologies. I see all of this causing edge devices that require low power consumption (like battery or solar powered devices) trending towards SigFox and LoRa as their primary connectivity approach.
Analytics on the edge
With cloud computing becoming more powerful and accessible, there has been a trend towards having more computing done in the cloud. While this allows heavy computing tasks to be offloaded from the edge device, it requires high bandwidth connectivity that’s highly robust to meet any real-time tasks, which is rarely available. The demand for edge devices to perform more complex real-time tasks (like image feature extraction) is causing a parallel trend in more analytics being computed on the edge. Processor manufactures are feeling this demand and responding with high performance, low power system on module (SOM) solutions that meet the lower price points needed for many edge devices. I’m expecting this convergence to continue and likely get to the point where demand drives the cost of higher performance SOMs below traditional lower performance microcontrollers.
It’s no surprise that with the explosion in the number of connected edge devices around the globe, security is becoming a huge issue that needs to be addressed. Many edge devices can’t support standard security technologies like TLS and SSL that have been developed for computer networks, so their data in many cases isn’t encrypted at all. To mitigate this, most communication protocols (like Bluetooth, Zigbee, SigFox, etc.) have implementation support for encryption, however, the commissioning of edge devices with encryption presents other challenges, so it still isn’t widely used. One of the trends caused by the increase in processor capability on the edge is the use of more standard security methods. Another trend is that sensitive data collected by edge devices (like GPS data) won’t be passed to the cloud but rather processed locally.
As AI and IoT technologies continue to evolve, some of these trends will either continue to grow or there will be new trends leading the industry. What trends do you see emerging in cloud computing right now? We welcome the opportunity to discuss your thoughts, or if you’re interested in learning about the 5 steps companies should take to monetize your IoT strategy, check out our eGuide here.