YOU ARE AT:Internet of Things (IoT)Seven elements of enterprise IoT device design

Seven elements of enterprise IoT device design

Design has been called “a formal response to a strategic question.” In an IoT context, the device is the embodiment of what strategic questions a business is considering as it pursues its use case, from cost strategy to security, to plans for what-if scenarios and how the device will be decommissioned at the end of its life.

Enterprise IoT device design must ensure that a device works properly and adheres to constrains on cost and aesthetic considerations. But here are some of the other factors for enterprise IoT devices:

Battery life. Wi-Fi company Edgewater Wireless recently supported an IoT deployment within Kroger stores, and Edgewater CEO Andrew Skafel said that battery longevity was one of the most important design factors. In a grocery store, there can be hundreds of sensors monitoring the state of food refrigeration, he noted, “so a couple extra years of battery life is a massive cost savings for them. I know that was front-and-center for their design.”

Connectivity choice. On this point, there is consensus: when it comes to IoT connectivity, there will be no consensus.

“No single standard will dominate the world,” said Mike Eftimakis, IoT product manager for chip company giant ARM. “It’s likely to be a combination of different solutions, different standards – and in in fact, we talk a lot about wireless but we shouldn’t forget that a good portion of IoT will probably be wired as well.” Some experts say that connectivity is actually one of the easier choices, because an enterprise’s business case is likely to be clear about whether short- or long-range technologies and licensed or unlicensed options are the best fit. Businesses are also choosing multiple connectivity technologies in their IoT deployments, not just one. Vodafone’s IoT Barometer report found that even small deployments of around 100 devices typically rely on three connectivity types to ensure the level of resiliency and coverage that they want.

Environment. Does the device need to function in extreme heat or cold? Stand up to vibrations? Can it be plugged in or does it have to rely on a battery? Operate on an enterprise’s own network, or someone else’s? The stability of the device’s deployed environment is an important factor in its design, said Terea Bui, who heads up IoT product marketing, strategy and go-to-market at Cisco. She pointed out the huge logistics differences between supporting a device inside of an office building versus an oil and gas or mining setting.

Application. Application development might be primarily a function of software, but it also has device implications in terms of the processing power and memory needed, and the extent to which a device must communicate with other devices or the cloud. James Kirkland, chief architect for IoT at Red Hat said Red Hat is seeing IoT move into open source, much the way that the cloud ecosystem has, because maintaining custom stacks was both too expensive and too difficult to manage.

“You have to look at the underlying technology and at your application and design it the same way. You can’t have single points of failure where it’s not scalable. Every point in the chain has to be something you can scale vertically or horizontally,” Kirkland said. He added that for some businesses, they design one aspect of the app to be huge but some back-end legacy service that has been the same for decades ends up as a chokepoint with a company not knowing how to scale, or being unable to scale, that piece. “It’s a tough challenge,” he said. “You have to understand the rate of growth, of that volume – five, 10, 20 years down the line. These applications have very long lives,” he added, and serving the enterprise means that businesses with expect extended device support, possibly for a decade or more.

Deployment. Physical design of the IoT device has to take into consideration how quickly and cost-effectively it can be deployed. This may not seem like a significant factor for a small-scale pilot, but it becomes vital at the point of going to market at scale, according to Von Cameron, VP of the Americas and global logistics for Actility. For the internet of things, the sheer manpower that it might take to put a device on thousands of “things” impacts form factor and deployment cost, Cameron said. “You need to think through, physically, how it’s going to get done,” he said. Can a device be simply stuck onto another object? Does it need to be plugged or screwed in? Does it take two hours or 10 to complete one installation and what level of skill is required? “The one thing you want to do is, once you deploy it, never touch that thing again,” Cameron said.

Operational costs. Cisco’s Bui said that one of the things commonly overlooked by companies new to IoT is the cost of operating the deployment once it is up and running in the field. Those costs include network transport, but also factor in support costs for remote devices that may can range from a few thousand in one market to millions across several geographic markets. She also said that a company needs to account for the costs of not executing on deployments plans – as in, what if a deployment that was supposed to take six to eight months drags into 12 to 18 months? That means months of missed revenues. Bui also noted that the device’s deployed environment will impact support costs as well, because of how easy or difficult it is to access the devices. The third area of operational costs are support: the size of the team necessary to support the deployment and how quickly tier one and tier two issues can be resolved, particularly in cases where a company intends to launch at scale across multiple markets.

Device lifecycle. When an enterprise is eager to create an IoT device and get it out in the field, sunsetting it may be low on the priority list. But Ashish Parikh, VP of IoT platforms and services for Arrow Electronics, noted that in order to have a comprehensive IoT design, the question of how to “responsibly and sustainably remove the device from service” is a part of long-term planning.

“Most people aren’t talking about this. Most people are saying, ‘I’ve just got to get it out there,’” he said. “But if you’re going to have 50 billion devices out there, what happens to all the electronics? What happens to the customer data sitting on the device? They need to be decommissioned at the end of their life in five years or 10 years out or 15. It still needs to be designed in and thought through.”

 

Looking for more information on enterprise IoT security, testing and design? Check out RCR Wireless News’ recent editorial special report and webinar.

 

Image: 123RF Stock Photo

ABOUT AUTHOR

Kelly Hill
Kelly Hill
Kelly reports on network test and measurement, as well as the use of big data and analytics. She first covered the wireless industry for RCR Wireless News in 2005, focusing on carriers and mobile virtual network operators, then took a few years’ hiatus and returned to RCR Wireless News to write about heterogeneous networks and network infrastructure. Kelly is an Ohio native with a masters degree in journalism from the University of California, Berkeley, where she focused on science writing and multimedia. She has written for the San Francisco Chronicle, The Oregonian and The Canton Repository. Follow her on Twitter: @khillrcr