beckmarkwith - Fotolia
When the Utah Division of Wildlife Resources was unable to handle the mounting piles of data it was collecting about the state's wildlife population, it turned to analytics tools from Google Cloud.
Data has long been used to monitor and maintain wildlife populations.
Wild animals are caught, tagged with tracking devices, set free and subsequently observed. The geospatial data collected then enables biologists, conservationists and others to study the habits of wild animals as well keep tabs on the population of animals in the wild.
Until May 2020, Utah's Division of Wildlife Resources (DWR) was using analytics applications developed and deployed on premises to monitor and maintain the wildlife population in the state's vast open spaces that include such national parks as Zion, Bryce Canyon, Arches and Canyonlands. The wildlife includes bison, cougars, elk, moose, mountain lions and wolves.
At first, it worked fine. But as more animals were tagged and more data collected, the system began to fail. By the spring of 2020, the Utah DWR's legacy wildlife tracking application was overwhelmed.
It simply couldn't handle to the volume of data being ingested and run queries on that data in a timely fashion, according to Eric Clark, account director for Google Cloud Services at SpringML, a consulting firm Utah's DWR turned to for help redesigning its analytics stack.
With SpringML's guidance, the Utah DWR ultimately decided to completely rebuild its analytics stack, and do so using the Google Cloud Platform.
Using Google's analytics platform would enable the Utah DWR to manage the amount of data it was collecting, according to Clark.
Perhaps more importantly, given the compute power of BigQuery -- a serverless cloud data warehouse from Google that enables analysis over petabytes of data -- also ensure it would be able to deal with the inevitable increase in data volume that will come as advances in technology enable even more data to be collected about wildlife.
"It's all about getting your data into a spot where you can start to take advantage of tools and features, and that's why we love using BigQuery for the centerpiece of a lot of the solutions we build," Clark said in a presentation during Google's Data Cloud Summit, a virtual user conference hosted by the tech giant.
Eric ClarkAccount director for Google Cloud Services, SpringML
With respect to Utah's DWR, he added that growing volumes of geospatial data played a key role in the decision to use BigQuery.
"Given the anticipated growth and the long-term needs of the DWR that really need to rely on this data, we really felt this was the best route in order to future-proof that solution and really position it for scale," Clark said.
By September 2020, Utah's DWR was equipped with a rebuilt wildlife tracker running on Google Cloud. At the front end is an application built with the Google App Engine, and supporting it is BigQuery.
"Now, we can focus on the data and the functionality of the application, not have to worry about the servers, and as the data grows and the user base grows as well, it's going to scale and meet the demand," Clark said.
As an example of what Utah's DWR can now do with data using Google analytics tools, Clark ran a query to see where a lone cougar tagged with a GPS tracking device traveled over a one-year period. Over that time, 5,300 location points were collected.
The query response was instantaneous, and revealed a travel pattern from east to west through the mountains of Utah. Applying a heat map, it revealed that the cougar stayed longest in the northwest region of its travels.
"Running a query with 5,300 points, the prior system would have struggled," Clark said.
A subsequent query to view the movement of all animals being tracked by the Utah DWR over a one-month period in the Book Cliffs mountains in eastern Utah contained about 63,000 location points. That information also was returned in under a second.
"That query never, ever would have worked," Clark said. "We've completely removed the constraint around performance."
Utah's DWR had a specific use case. That use case, however, exemplifies what any organization can do with data if they manage it properly and position it for action, according to Clark.
Quoting statistics from IDC and Seagate Technology, Clark said that organizations worldwide are predicted to create 175 zettabytes of data by 2025. At the start of 2020, it was estimated that there were 44 zettabytes of data.
Every organization is going to have to deal with the same increase in data volume Utah's DWR is now managing, and every organization is going to have to employ an analytics stack capable of handling an exponential increase in data.
That begins with where the data is stored and managed, according to Clark.
"When we think about modern analytics and getting the most value out of the data you already have and will have in the future, it turns out that the position of your data really matters," he said.
It will determine whether organizations can do advanced analytics including predictions and forecasting, whether they can adopt machine learning and augmented intelligence capabilities, can consolidate and connect various sources of data, and can future-proof themselves, Clark continued.
"Thinking about the amount of data growth we know we will see in the coming years, we need to be thinking about to grow our capabilities around managing data," Clark said. "We need to be thinking about how we can analyze it and not have to think about the infrastructure, so we can put as much attention as possible on the insights."