Kinetica on NVIDIA DGX Station Helps Rescue The Honey Bee!

My daughter, a high school junior, came home one day talking about her upcoming science fair, and how she was interested in addressing a serious problem with the domestic honey bee. Apparently, domestic honey beehives are collapsing at an alarming rate due to parasitic mites called Varroa, and she provided a theory that a healthier bee habitat design may be a more natural approach to fending off parasites and disease than the current failing system. A healthier organism overall makes for a stronger, more natural defense against these hive-collapsing pests. She even had a plan to build an AutoCAD design from her research on a specific natural hive design, called The Sun Hive, and potentially produce a 3D printed prototype. Cool, right??? I thought so too and was interested to see how it was all going to take shape in the coming months. So, I mentioned it to a couple of colleagues in passing.

Jade Greenberg working on 3D CAD design

Ben Siegel, CSUSM CS Major, Jacci Cenci, Nvidia SA at The Ecology Center in San Juan Capistrano, CA

As a business analytics enthusiast and practitioner, my water cooler discussions turned to different data collection strategies with sensors, and some of my technical peers offered some interesting ideas. Many of Kinetica’s technology partners bring a wealth of knowledge and experience across many interesting domains, but I never imagined that a casual conversation with one of them around my daughter’s interesting beehive project would have influenced the chain of events that followed.

Enter Jacci Cenci, Solution Architect with NVIDIA, and former NASA scientist by the way, who proceeds to tell me about some new NVIDIA edge computing technology that can potentially use live video to identify unhealthy bees in the field in real time. Inspired by the concept of connecting these advanced technologies with a high school student, she took the ball and ran with it! Not only did she start connecting with her technology partners to discuss the machine learning development, she also offered to be my daughter’s mentor for her science project. Now my daughter had real-world support to help realize her vision. With exposure to the latest intelligent monitoring technologies available, she can conceptualize cutting edge methods for measuring her beehive design performance in the field. With these types of analytic insights, and by leveraging AutoCAD design tools, along with modern 3D printing for rapid hive prototyping and production, the evolution of her design can be accelerated like never before!

3D printing in action

In addition to mentoring my daughter through her science project analytics, Jacci managed to connect with a honey bee farm scientist in California who agreed to field test this real-time monitoring solution with her beehives. As Jacci’s field efforts progressed, from developing video algorithms to collecting sensor data, the necessity for a real-time data collection, machine learning, and analytics visualization created an opportunity to introduce Kinetica. Now I was able to contribute, and so I engaged a few of my Kinetica colleagues to help round out this real-world environmental analytics opportunity. Ultimately, the Kinetica and NVIDIA teams were able to collaborate and demonstrate how new accelerated analytics technologies, and their corresponding compute platforms, can deliver game-changing possibilities for innovation. This was all started by a high school student’s science fair project!

When you step back and look at all of the technologies we have been able to apply as a team to this challenge, across organizations, in relatively short order, it demonstrates to me how adaptable and accessible these complex, cutting-edge technologies have become. The disciplines applied in this case included:

• IOT streaming sensor data in real time
• Edge computing solutions that process image inferencing algorithms on still images and live video in the field
• GPU accelerated database engine for real time data ingestion, query, machine-learning, and visualization

The entire Kinetica solution deployed to an NVIDIA DGX Station, ultimately accelerating our productivity by running everything on one, GPU-enhanced platform.

Next step? Honey bees to the rescue!

NVIDIA DGX Station

Kinetica presented this solution at the O’Reilly AI conference earlier this year which you can view here:

Jade was awarded 1st Place for Engineering at the 2018 Nokia Bell Labs New Jersey Regional Science Fair and selected as a finalist to attend the 2018 Intel International Science and Engineering Fair.

You can watch Jade talk more about her beehive design and rescue strategy here!