4 steps to take when implementing an IoT stack for integration

Using an IoT stack can help your Internet of Things hardware work together smoothly. But there are things to consider before adoption.

industrial-iot.jpg

Image: iStockphoto/ipopba

Choosing the right Internet of Things stack can be an effective way to implement IoT in an organization if the stack can not only integrate IoT communications and data flows, but also enact uniform security policies for the entire stack. 

IoT solution providers know this: They understand that companies are struggling with IoT integration. If you can provide an end-to-end IoT solution consisting of hardware, software and a set of development tools that IT can use at each level of the stack, you can come a long way in offering IT ease of IoT integration.

SEE: TechRepublic Premium editorial calendar: IT policies, checklists, toolkits, and research for download (TechRepublic Premium)

Of course, if you are IT, the task isn't that simple. You still have legacy applications and downstream data marts for data analytics that you will need to integrate with the IoT stack. There are also literally hundreds of IoT vendors and solutions in the market, with no guarantee that every single IoT device your company deploys will neatly integrate into the stack.

Nevertheless, there are several steps you can take to utilize an IoT stack approach within your larger enterprise integration strategy.

1. Understand what an IoT stack will and will not do for you

In some cases, the IoT stack might favor certain IoT technologies and even proprietary technologies. The best choice is an open IoT stack solution that will fit long-term with your company's technology direction.

2. Appreciate the difference between integration and interoperability

If IoT in the stack integrates well at all levels of IoT, it means that to a degree, you can "plug and play" different types of IoT devices, and that these devices will be able to exchange data through a uniform set of application programming interfaces that the IoT stack provides. A comprehensive set of APIs reduces IT's need to hand code individual interfaces. 

SEE: 5 Internet of Things (IoT) innovations (free Pdf) (TechRepublic)

If the IoT is interoperable and not just integrated, it means that the IoT devices, gateways and platforms actually use the same data formats for the data being passed. Since data is uniform, this simplifies the later job of having to clean and normalize data for each IoT device separately, since the data already is uniform if it is truly interoperable. With interoperable data, there is only one IoT data input that has to be cleaned and transformed so it can integrate and interact with other downstream data from your legacy applications.

Most IoT data is of the integrated variety—meaning that you likely will have the bigger job of separately transforming the data from each IoT input into your central data repository, instead of just having to do this job once.

3. Vet IoT security in the stack

There are a plethora of security issues at the IoT device level, as well as on IoT gateways and platforms. The IoT device level can be especially troublesome, since device vendors can add security to settings that don't conform to corporate security standards. In these cases, IT must manually adjust and test the settings. When you interview potential IoT stack vendors, ask them about their ability to apply uniform security policies and standards at every level of the stack and how they do this for IoT devices that you purchase on the open market.

SEE: Rogue IoT devices are putting your network at risk from hackers (ZDNet)

4. Have a downstream data plan in pace

You might have a very effective IoT stack, but this won't take care of the downstream data architecture and integration issues that arise when you want to combine IoT data with other data from legacy systems in a larger data repository. The ultimate compilation of data from all points of the enterprise into a single data repository is still an integration job that IT must plan for.

Why it's important

IT professionals know that system and data integration are some of the most time-consuming and unappreciated jobs—but that integration is all-important if your company is going to bring together data from disparate systems and operate on it.

What the IoT stack offers is an end-to-end solution that can simplify the integration of IoT. Because it does this, IT should highly consider a stack approach to IoT—with the caveat that additional downstream integration to other systems will still be needed.

Also see