AI-powered research: Oregon State University’s work at sea

Artificial intelligence is accelerating scientific research, allowing for deeper insights at an unprecedented pace. At Oregon State University’s College of Earth, Ocean and Atmospheric Sciences, AI-powered research is driving new discoveries in oceanic and earth sciences, transforming how researchers process vast amounts of unstructured data.

Dell Technologies Inc.’s advanced infrastructure is powering this research, enabling real-time analysis of critical environmental changes, according to Christopher Sullivan (pictured, right), director of research and academic computing at CEOAS.

“We really do focus on how the planet is being changed by the things that we do and the way we interact with it,” Sullivan said. “We use a lot of data, obviously, to help us do that.”

Sullivan and Dell’s Joe Steiner (left), principal system engineer, and Anthony Dina (center), director of system engineering, spoke with theCUBE’s Dave Vellante for the “Oceans of Data: Tackling Climate Research With Data and AI” event, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed how AI, real-time processing and scalable infrastructure are shaping the future of environmental research. (* Disclosure below.)

Unlocking ocean science with AI and high-performance data

CEOAS’ research spans ocean and land, studying everything from plankton and seabirds to freshwater ecosystems and carbon cycles. These projects generate immense datasets, requiring advanced AI models to extract meaningful insights. Dell’s technology plays a key role in enabling AI-powered research by providing scalable infrastructure, according to Sullivan.

“Ultimately, we have computation, and we just have a gamut of data coming through,” he said. “Before we had AI … we were data-rich and information-poor. One of the greatest advents for AI was the chief hard drive. It was about data. AI is data-driven, and it’s married to that data.”

Processing data efficiently is just as critical as collecting it. CEOAS researchers rely on specialized infrastructure tailored to the demands of different scientific questions and configure machines with varying capabilities to meet diverse computational needs, unlike traditional enterprise environments that standardize hardware, according to Sullivan.

“One of the great reasons we use Dell is because we get the ability to build out different platform technologies that really focus in on the questions we’re trying to answer,” Sullivan said. “Different data requires different technologies for processing. I actually have to work with Dell to configure them to actually do the science. If I can’t deploy the technology in an array of different ways, we won’t actually be able to do the science.”

Taming data complexity for AI-powered research

CEOAS’ AI-powered research generates massive datasets that must move rapidly between disciplines. The university has built a high-performance on-premises infrastructure optimized for speed and efficiency to support this need. By leveraging Dell’s advanced computing platforms, CEOAS accelerates AI-powered research, allowing scientists to extract insights faster than ever, according to Sullivan.

“We have too much data to be putting that [work] in the cloud and getting that work done; we need to have on-prem, and we need to own that on-prem and move it around between our different disciplines and domains very rapidly,” he said. “Dell really did help us recently … [to] change the way that we interact on-prem with our data by helping us look at 800 gig networking between our data centers. That really changes the way that we can interact with our data — we are talking about petabytes.”

Beyond infrastructure, structuring unstructured data is another major challenge. AI models rely on metadata tagging and annotation to describe images and sounds, making information searchable and actionable. At the same time, governance and security are critical for ensuring proper access controls and mitigating cybersecurity risks, according to Dina.

“There’s been a huge demand on providing structure to unstructured elements,” he said. “Not only do we need to make sure the right people have access in the process, but we have to prevent the intrusion from cyber-attacks. We have to think about chain of custody in a whole new way — not just at a file level, but at a data level.”

AI workflows also depend on efficient data handoffs. Instead of starting from scratch, researchers build on pre-processed data, reducing costs and accelerating insights. This approach enables teams to collaborate effectively while ensuring data governance remains intact, according to Steiner.

“I think about a relay race,” he said. “I think about how everyone may run a distance and then hand off information, and the next guy is able to run extremely hard. We’re not starting from ground zero anymore. The base information’s already been crunched on, and now we’re able to pick that up and build on top of that at an accelerated pace.”

Bringing AI to the ocean for real-time insights

For CEOAS, real-time data processing is a game changer. Research vessels, including the newly outfitted Regional Class Research Vessel Taani, are equipped with Dell PowerScale and PowerEdge systems, as well as Nvidia GPUs, enabling AI-driven analysis at sea. Instead of waiting until they return to shore, researchers can now process data as it’s collected, accelerating discoveries, according to Sullivan.

“If we use that [plankton] as a monitoring tool, it’s really like the canary in the coal mine to helping us understand how we’re changing the planet,” he said. “We really want to be doing it in real time because if it takes us too long, the data becomes meaningless.”

Dell’s and Nvidia’s infrastructure also powers advanced marine ecosystem monitoring. Using autonomous robotic vehicles outfitted with 8K cameras and AI models, researchers analyze billions of plankton in real time — identifying rare species and tracking oceanic health with unprecedented precision. AI’s ability to rapidly mine and reprocess vast datasets allows scientists to uncover previously undetectable patterns. From real-time data collection at sea to advanced marine ecosystem monitoring, AI-powered research is reshaping how scientists study the planet, according to Sullivan.

“There was no way you could possibly have a human quantify that,” he said. “Statistically, we’re actually making relevance now, and I have a lot more ability to see the rare species that are not out there in heavy abundance because we’re able to mine through it so rapidly. We’re towing at five knots. We are able to do 162 liters a second.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE Research’s coverage of the “Oceans of Data: Tackling Climate Research With Data and AI“ event:

(* Disclosure: TheCUBE is a paid media partner for the “Oceans of Data: Tackling Climate Research With Data and AI” event. Neither Dell Technologies Inc., the sponsor of theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE