Hannah Parraga transforms big data for big impact

Big science requires big data. Without the in-house computing expertise at the U.S. Department of Energy’s (DOE) Argonne National Laboratory, processing the exceptionally large amounts of data collected by experiments at the Advanced Photon Source (APS), a DOE Office of Science user facility at Argonne, would be impossible.

Software engineer Hannah Parraga is a member of the Scientific Software Engineering & Data Management Group (SDM) at the APS, providing that expertise. Her work focuses on transforming raw detector data into usable formats for researchers so they can make scientific breakthroughs.

Argonne accelerates the science and technology that drives U.S. prosperity and security. Argonne staff are thinkers, builders, problem-solvers and caretakers — every person and every role contributes to the laboratory’s impact on the world.

“I often click through the publications list on the APS website. Seeing the breakthroughs that happened here reminds me why this work matters. Advances in medicine, energy storage technology — not to mention improvements in products we use every day, like a new toothpaste formula — are realized at the APS.” — Hannah Parraga, software engineer at the APS

Researchers come to the APS to investigate all kinds of materials using a myriad of X-ray techniques from tomography to ptychography and beyond. To handle this breadth of scientific inquiry, the APS houses 72 X-ray beamlines to meet differing needs.

“Each APS beamline has its own way of working, and they all have different data requirements,” Parraga explained.

As part of the SDM group, Parraga maintains the APS data management system. When the raw data begin to come in, she automates the process so that pieces of data are integrated. She also ensures that the data gathered at the APS remains accessible and secure after researchers leave the facility.

The recent APS upgrade presented new opportunities for scientific discovery but some challenges for computing. Especially challenging are the skyrocketing data volumes from accelerated performance.

“Before the upgrade, we could process data on local computers,” Parraga said. ​“Experiments on the new beamlines can generate terabytes of data throughout the day. Processing this unprecedented amount of data requires powerful computing resources.”

A terabyte is a unit of digital information equal to 1,000 gigabytes. To put this in perspective, the largest cell phone models currently available can store up to 1 terabyte of data, which means they can hold roughly 250,000 photos, or hundreds of hours of high-definition video. Older beamlines typically generated data on the order of gigabytes. Upgraded beamlines will generate significantly larger amounts of data.

That’s why the APS has partnered with the Argonne Leadership Computing Facility (ALCF), another DOE Office of Science user facility located at Argonne.

“The ALCF allows us to run APS data processing tasks in an on-demand fashion on their supercomputers, so that computing is ready when APS experiments need it,” she continued. ​“This collaboration has been crucial in meeting the demands of modern science.”

Much of Parraga’s recent work has involved writing data processing pipelines that take advantage of the supercomputers at ALCF to handle these massive datasets. This effort is part of Argonne’s Beamline Data Pipeline Project, a collaborative mission to use the capabilities of the upgraded APS.

Parraga works closely with APS users to understand their daily operations and evolving needs.

“Getting direct input from the scientists helps me tailor solutions that really support their research,” she said.

Staying connected with the science happening at the APS also keeps Parraga inspired.

“I often click through the publications list on the APS website,” she said. ​“Seeing the breakthroughs that happened here reminds me why this work matters. Advances in medicine, energy storage technology — not to mention improvements in products we use every day, like a new toothpaste formula — are realized at the APS.”

It’s important to Parraga that her work have a positive impact on the world. She’s happy she found a place that supports, in her words, ​“really great and transformational science.” Argonne’s Core Value of impact particularly resonates with her, she said.

Since joining the lab over four years ago, Parraga has earned a master’s degree in data science from an online program through the University of Wisconsin–Oshkosh. Her colleagues have supported her the whole way.

“At Argonne, I’m surrounded by brilliant people who are leaders in their fields,” she said. ​“Many of my colleagues have decades of software engineering experience, and I’ve learned so much by working alongside them.”

Parraga shares her own expertise by participating in volunteer initiatives at Argonne, including fundraisers for scholarships and the Coding for Science Camp for high school students.

“Argonne’s commitment to community outreach is wonderful,” she said. ​“I appreciate all the programs for students and the public — from CyberForce competitions to OutLoud lectures and open houses.”

Parraga can personally attest to how programs like these can change a young person’s life. Her passion for programming was kindled during a college hackathon.

“That was my real ​‘aha’ moment,” she recalled. ​“Seeing people come together, brainstorm and build projects over a weekend showed me that the sky’s the limit with programming. You can create anything you imagine.”