Three Best Practices for Navigating the Internet of Things

Kevin Petrie's picture

We’ve all seen enough science fiction, and new gadgets coming to market, to envision the approaching world of smart homes, cars and appliances that can automate our personal lives.

In reality, the Internet of Things (IoT) holds perhaps the greatest potential for corporations. 

The concept is not new – telemetry, for example, arose in the 19th century.  But widespread broadband and new narrowband options, increasingly small and cheap sensors, and new machine-to-machine connectivity are creating powerful possibilities for organizations to fine-tune or even revamp their business.

The IoT is an opportunity, but it’s also a mandate.  Manufacturers, retailers and other companies whose core business involves producing or handling physical objects must adopt IoT to compete effectively over the next 5-10 years.  According to Baseline Magazine, about two thirds of corporate IoT adopters seek to reduce cost, 22% to manage risk, and 13% to innovate or grow. 

The IoT will force substantial changes to how IT works today.  Reasons for this include the following:

  • Data containers (i.e., software-based digital files) are proliferating, in part because RFID tags and other sensors have so few bytes.  The EMC/IDC Report on the Digital Universe noted in April 2014 that “the number of “containers” is growing faster than the number of petabytes, from 28 quadrillion in 2010 to 4,200 in 2020.”  Consolidating the data within them will require new skills and technologies.
  • The time value of data will be even higher, prompting the need for faster, even real-time analytics. IoT is based on sensors that capture moments in time: consider factory pressure gauges, traffic patterns or point of sale transactions by an identity thief.  The real-time imperative means that organizations must quicken and ideally automate their data loading and analytical processes.
  • The Internet of Things extends beyond the cloud or data center.  Data collection, analysis and the resulting actions will require close teamwork with field technicians or other stakeholders that might not have engaged IT professionals in the past.

So how should IT practitioners and management approach IoT?  While this trend cuts across verticals and geographies, here are three standard best practices to follow.

  1. Base your IoT strategy on corporate strategy and not vice versa.  This mantra, used in many other technology discussions, bears repeating.  The temptation for departmental science experiments can run high given the hyperbole surrounding IoT.  But as with Big Data and opportunities to innovate, business objectives and strategy should dictate the plan for IT.  If a business unit is going to test an IoT theory, they should do so with a decision tree and funding that executives have blessed.
  2. Invite your line workers to brainstorm.  More than ever you’ll need the insights of staffers that have a variety of technical insights about how your company can play sensor-based data to competitive advantage.  Mechanical engineers that run oil rigs and other heavy equipment might not be typical partners of IT, but still have the best ideas about how to combine data to improve operations.
  3. Use the most flexible repository possible, and invest in mobility.  New IoT data sources and formats are harder to handle.  Companies will need more flexible repositories to be able to extract intelligence from all the pieces.  Hadoop Data Lakes can be a good starting point because Hadoop can receive data in its raw format.  But Hadoop has some limitations, as Michael Hausenblas with MapR points out in a recent article in DataInformed, and some analytics exercises will require distinct repositories.  Work closely with your IT organization to align your target data, their formats and your analytical goals.  And shop for software solutions that can automatically move data across a wide range of platforms.

Established tech vendors are investing heavily to help enterprises navigate the complexity of the IoT. SAP announced this week new HANA Big Data Intelligence capabilities for enabling organizations to rapidly ingest and analyze data from a variety of sources, using the HANA platform with its Event Stream Processor, SAP IQ software and Apache Hadoop. While HANA’s in-memory processing is a natural fit for in-line analytics, SAP clearly understands the need to embrace new IoT data sources and platforms.

As you invest in IoT, focus on corporate strategy, enlist your broader workforce, and stay flexible with your infrastructure.

To learn more about IoT solutions, download the Attunity whitepaper, Releasing the Value Within the IIoT.


Download the Whitepaper