Does the Internet of Things Always Mean Big Data?

In an interesting article entitled Forget Big Data -- Small Data Is Driving The Internet Of Things, Forbes writer Mike Kavis makes the point that it’s inaccurate to assume that the Internet of Things necessarily presupposes big data. He’s right in an obvious way, to get to big data you need “small data,” but I don’t think it’s reasonable to assume that for the majority of Internet of Things applications, big data is is unimportant.

He defines small data as “a dataset that contains very specific attributes. Small data is used to determine current states and conditions or may be generated by analyzing larger data sets.”

The obvious objection to the line that small data is at the heart of the IoT is that if small data is generated from larger data sets, you need big data to get at small data. It’s a commonplace in the big data world that the useful information contained in data lakes has to be extracted from the mass of data through analytics, and that useful information may not itself constitute big data — it may simply be a handful of plot points on a graph. But that’s not where the real problem lies with the claim.

Kavis goes on to say:

“Optimizing these business processes can save companies millions of dollars through the analysis of relatively small datasets. Small data knows what a tracked object is doing. If you want to understand why the object is doing that, then big data is what you seek. So, the next time someone tells you they are embarking on an IoT initiative, don’t assume that they are also embarking on a big data project.”

It’s true that it’s useful to know what an object is doing, but business is wasting an opportunity if they decline the big data benefit of knowing why it’s doing it.