Brian Chee, University of Hawaii at Manoa

Brian Chee
Interop 2007 Transcript

00:00
Hi, I’m Brian Chee, I’m with the University of Hawaii’s School of Ocean, Earth Sciences and Technology.

00:14
Splunk is kind of a unique product. It allows us to take a lot of time-series data. Now, if anyone’s ever played with sonar, or ocean observing, it’s all time-series. And one of our big challenges has been, how the heck do we organize, and how the heck do we mine this rich source of data?

00:31
Every time we sail out we sail out on one of our research vessels, we tow behind us a side-scan sonar, it’s like a torpedo, and we put it on a long tail, and trail it behind the ship.

00:42
We call it, "mowing the lawn", because it’s actually a strip about thirty kilometers wide, depending on the depth, and we map the sea floor. Well, getting to all this data has been a continuous challenge for us, and we’re always looking for new ways of organizing and mining that data.

00:59
Splunk is one of the solutions that we’re taking a good, hard look at because of its value in being able to organize it in the form of a timeline, and line up events from multiple data locations.

01:10
So, we might have an undersea microphone; in this case I have one that’s one hundred kilometers north of the Hawaiian Island chain, and five kilometers deep. We’ve got, in the summer, whales. We’ve got krill clicking in the background. We’ve even heard a Chinese diesel-electric sub go by.

01:30
There’s all these wonderful events that are all time-series data. But how do you correlate them together? How do you organize them?

01:37
In this case we’re taking a very hard look at Splunk. And, who knows? Maybe one of these days you’ll be able to go to our website, issue a command, and ask for all the data from a certain spot in the ocean, organized by Splunk.