Vivint Drives Smart Home Automation With Datameer

Quickly and easily integrate data streams into Hadoop

Join and analyze the data streams with Datameer’s easy-to-use, spreadsheet-like interface in minutes

Parse, join and sessionize various complex data streams

Vivint

Vivint Smart Home, a leading smart homecompany in North America, is leading the charge in the data-driven, connected home movement. Serving more than 1.5 million homes, Vivint’s touchscreen panel, the hub through which all of their other products communicate, creates a streamlined network thatconnects all of the home’s smart systems including security, HVAC, lighting, small appliances, video, and other devices.

Industry

Home Automation

Use case

Customer Analytics

As a part of the Internet of Things and big data, Vivint’s solutions enable homes to be smarter, safer, and more energy efficient forits customers. Intelligence is the center of the smart home and Vivint is using Datameer to shorten the time it takes to go from raw data to actionable intelligence that can be used to enhance their customer’s home experience.

Vivint had selected Hadoop,an open source storage and compute technology, as their analytics infrastructure platform. They found that their team was spending too much time on mundane, technical tasks preparing and integrating the data rather than doing actual, value-added analysis.Vivint was looking for a solution that could make their staff more efficient and seamlessly integrate with their Hadoop analytics platform rather than just retrofitted to work with Hadoop. Another key consideration was the ability to integrate and analyze not just raw data but also streaming data, which is a key component to their smart home analytics solution to big data for Internet of Things.

After reviewing a number of technology options, Vivint selected Datameer as their big data analytics solution. Datameer is used to join and analyze terabytes worth of data collected from its Internet of Things solution and various in-home automation, security, and energy management devices, including thermostats, smoke alarms, automatic locks and door and window sensors to better understand usage patterns
and improve services. A daunting task with traditional systems, Vivint analysts use Datameer to parse, join and sessionize various complex data streams to determine occupancy and vacancy patterns in an effort to reduce the number of false alarms and improve the overall efficiency of their devices.

“One of the challenges and opportunities with the Internet of Things is that there
has been an explosion of stream data, which is unlike the traditional data we’ve
worked with,” said Brandon Bunker, senior director of customer analytics and
intelligence at Vivint. “Instead of having one row per customer there may be
thousands or millions of rows per customer. Since Datameer is purpose built for
Hadoop and it has an intuitive Excel-like user interface, my team was able to get up
and running very quickly.”

Unlike most traditional business intelligence or data analytics tools that sit on top
of traditional databases and require data modeling, Datameer uses Hadoop as its
underlying storage and compute platform. This allows analysts to work with raw
structured, semi-structured, and unstructured data sources fluidly when dealing
with big data and Internet of Things. Because Vivint connected devices produce
unstructured stream data, analysts at Vivint are relying on Datameer to quickly
and easily integrate data streams into Hadoop and then join and analyze the data
streams with Datameer’s easy-to-use, spreadsheet-like interface in minutes.

Data-driven products that help homeowners manage their home safety and energy is
the future in home automation because it improves the overall homeowner experience.
Vivint continues to remain a leader in the industry and, together with Datameer, is
making strides to define the new status quo in the connected home movement.

Instead of having one row per customer there may be thousands or millions of rows per customer. Since Datameer is purpose built for Hadoop and it has an intuitive Excel-like user interface, my team was able to get up and running very quickly.

Brandon Bunker

Vice President of Engineering