The Internet of Things is experiencing a remarkably fast rollout given its size and complexity, but this is leading many enterprises to confront key challenges before developers have had a chance to work them out.
This is a testament to the faith that organizations have in the value of their data and the ability of advanced collection and analytics capabilities to unleash that value in a tangible way. But just as drilling for oil is still a hit-or-miss game despite the advances geologic surveying, so too will the enterprise encounter difficulty finding the real data amid all the junk.
As I mentioned a few weeks ago, basic connectivity is likely to be an issue, particularly when dealing with consumer-facing applications. Many of the leading IoT use cases require real-time streaming data and analytics, which is compromised in the presence of multiple network protocols spread across the universe of connected devices. As well, many of the large processing centers, or data lakes, that will churn through the bulk of IoT information will likely incorporate a variety of vendor solutions that ideally should communicate across a language, the beginnings of which are only starting to emerge.
At the same time, IoT infrastructure will need to be integrated into legacy back-office systems if the enterprise hopes to utilize this wealth of new data for existing, or entirely new, data processes. According to Machina Research, 38 percent of enterprises are already using IoT solutions in one form or another, and that level is expected to grow to 81 percent as early as 2018. By the end of the decade, 43 percent of IT budgets will be consumed by IoT initiatives, with a good chunk of that devoted to integrating the technology into legacy operational systems, resource management functions and business applications like CRM.
Some developers are already gearing their products toward broad IoT functionality.