by Minda Zetlin, August 2014
Businesses large and small will derive huge benefits from the wave of machine data and other information the Internet of Things (IoT) provides. Here are some issues to consider before starting an IoT project.
“In the past, people have just said, ‘We’ll store the data, and then we’ll figure out what to do with it,’” says Andy Piper, CIO at Push Technology, which makes data distribution software. “The IoT breaks that model. The idea of storing data and figuring out what to do with it later is not a viable option.”
The best solution, he says, is to automate the data selection process, using technologies such as complex event processing. It’s important to do this as early as possible, he adds, “so you’re only analyzing the data that’s relevant to you.”
What data to keep and for how long is another challenge. And there are timeline issues: some data loses its relevance very quickly, while other data should be kept for the long term because it provides valuable historical perspective. How do you know which is which?
“You have to know your business really well and be able to make those decisions quickly,” Piper says. “If you don’t have that context, you can’t.”
Security was IT leaders’ #1 concern about the IoT in recent surveys by Opinion Matters and VDC Research. The IoT presents an array of security challenges, especially because many small devices such as sensors don’t have much in the way of an operating system and can’t necessarily carry their own security. Thus the best place to secure IoT data may be at the collection point or the point where it’s about to be sent over the internet.
With IoT technologies controlling more and more machines and processes in the real world, it’s easy to imagine scenarios in which a malevolent hacker could gain access to systems and cause great harm.
A second concern is that many IoT devices are disposable, which means they could conceivably be discovered and read. “Decommissioning a throwaway device that’s measured my heart rate for the past year, do I want to put that in the landfill with the data about me in it?” asks Tim Murdoch, director, digital services, at Cambridge Consulting.
Dealing with the data that IoT devices can generate also requires very robust network bandwidth, storage, and processing power. In fact, improving infrastructure might be the necessary first step in an IoT project, according to Gaurav Palta, partner, high tech and manufacturing, at Infosys Consulting. “The biggest challenge a lot of companies have is that they’ve grown organically and over a period of decades have developed a complex ecosystem of databases and applications and integrations,” he says. Companies with this history should make sure to prepare for the IoT—and make sure that they are using the right tools to integrate existing technology.
Many of the things in the IoT run divergent operating systems with few standards among them. Experts agree that a standard will evolve over time—Oracle Java Embedded, which is open source and has a long history in the marketplace as well as deep market penetration—may be that standard for many types of IoT devices.
In the meantime, the protocols by which devices can share information over the internet take on more importance—along with the middleware that helps gather data and connect it to other devices. “A lot of companies are proposing middleware to connect our devices,” Murdoch says.
Multiple companies are fighting to dominate the middleware space, he adds. “Ultimately, we’ll reach a level of maturity where a few common approaches will dominate the market,” he says. But not just yet. “There will be a lot of companies with their own solutions for a long time to come.”
Minda Zetlin is coauthor, with Bill Pfleging, of The Geek Gap: Why Business and Technology Professionals Don’t Understand Each Other and Why They Need Each Other to Survive (Prometheus Books, 2006).