Using Times Series Data to Optimize Your IoT Business Objectives

To access the Big Data landscape and a recording for this event, visit the dashboard here.

Readwrite Labs had a masterclass two weeks ago in partnership with Influx Data to discuss data and IOT. We also drilled down on Time series data which is the major type of data involved in an IOT implementation. 

Today’s Internet of Things (IoT) environment is here to stay and shows every sign of becoming an integral part of business and consumer lives. Estimates put global IoT connected device use at 30 billion by 2020. This number is set to grow as more aspects of work and life rely on IoT.

Along with this growth, data is rapidly expanding. IDC’s white paper, “Data Age 2025,” noted that it’s now predicted an incredible 163 zettabytes of data will be produced around the world by 2025. To provide a better picture of what this means, one zettabyte is the equivalent of one billion terabytes or one trillion gigabytes. This new data estimate is significantly more than previously thought, illustrating the shift in how businesses and consumers see value in getting and sharing all types of information.

IDC’s research also uncovered a definitive change in who produces this data, with enterprises creating more than 60 percent of this data by 2025 versus consumers. That leads to both an opportunity and challenge for today’s organizations to determine how to manage data to their advantage.

A Wider and Deeper Context

Today, there are smart consumer and enterprise applications, smart cloud and network platforms, and connected and autonomous things. A massive amount of data is being created across all those platforms and needs to be transformed into business insights automatically so it can be easily consumable. It’s this complex connectivity that is driving businesses to transform the proliferation of data into actionable business insights. And, with such diverse IoT implementation needs, it’s difficult to find one solution that can handle it all.

With the rapid acceleration in data availability and the continual movement between connected points — businesses, consumers, cloud IoT devices, and networks — organizations need to synchronize data movement and management. To get the most from the available data, they must be fast and precise with aggregating, organizing, storing, and using this data.  

Data Architecture is the Answer

There is a need for order in how data travels and is used across the board. In the same way urbanism determines how people are going to live in a city and move from one place to the other, data architecture is critical to how data will be leveraged and consumed across the board. An organization must carefully consider the available data architecture to identify which one aligns with the unique traits of their IoT data. That data architecture must address all components of IoT data, including aggregation, analytics, visualization, and storage.

When properly implemented, a business can gain real-time insights, accelerate decisions, and meet — or exceed — customer expectations. Yet, more often than not, organizations don’t use the right data architecture, and they are not getting the most from the aggregate data, which stays idle like unmined gold.

What is Your Optimization Objective?

Where organizations often go wrong when selecting a data architecture is not starting with defining their optimization objectives. In beginning with the end result, it will narrow down the type of data architecture that works.

For example, the data architecture for satisfying customers is very different than what should be used for optimization objectives like safety or business insights. Additionally, there are other layers of insights that may be important to an organization from a management perspective or that of an operational viewpoint.  Once that objective is established, the next step is to examine the traits in the data being aggregated that will further direct what platform to use for the data architecture.

Using Times Series Data Architecture

According to Chris Churilo, Director of Product Marketing for InfluxData, the leader in times series platform, “What we see among organizations is that the majority of organizations rely on traditional relational databases followed by document databases and then time-series platforms. Still others are not sure what they should use to aggregate, analyze, and store their data.”

Times series platform are particularly suited to a data architecture that needs to address  what happens every second of every minute with an organization’s data. This provides a way to identify anything that is not working efficiently and make immediate changes to how the data is being aggregated and organized. Plus, it delivers the insights necessary to quickly make strategic pivots in order to address objectives like customer satisfaction and safety.

With InfluxData’s architecture, an organization can handle large volumes of data points per second while organizing real-time queries across large data sets. Its open source platform makes it easy to deploy, use, and integrate with other database technologies.

For example, InfluxData’s Telegraf™ now integrates with Google Cloud IoT to connect, process, store, and analyze data within an organization’s IoT environment. By using this plug-in for Google Cloud IoT platforms, which is one of over 160 plug-ins available from InfluxData, an organization can start collecting and analyzing sensor data within hours of implementation. Manufacturers using this platform can optimize for safety, including anticipating problems and improving operational efficiencies by leveraging time series data.

Additional Benefits of a Time Series Data Architecture

In reflecting on the new increased amount of data that will impact enterprises in the next few years, it’s critical that organizations look to data architecture solutions like time series now or risk losing this “gold.”

An open source time series data architecture means that it can be adapted to directly address an organization’s specific optimization objectives, including collecting a more diverse range of metrics. From there, an organization can effectively manage situations like predictive maintenance or optimized traffic routing.

Plus, an organization can implement additional automated and control applications for shutting down faulty equipment in a remote area without the need for any human interaction. Therefore, the IoT environment can be optimized, in this case, for safety. The same can be applied to other objectives, delivering customer satisfaction or furthering business or operational insights.

By having the right data architecture, an organization can collect this time series data to observe, learn, act, and automate for any system across all industries. Rather than worrying about the imminent arrival of significantly more data, enterprises can embrace it, knowing that their data architecture is aligned with the volume, complexity, diverse applications, and specific objectives of the IoT connected environment.

View the full resolution Autonomous Vehicles landscape.