This month, Helen Xu begins her role as Berkeley Lab’s new Grace Hopper Fellow. Established in 2015, this prestigious fellowship aims to develop young computer and computational scientists to make outstanding contributions to high-performance computing (HPC) applications.
Xu said she’s excited to bridge the gap between computer science and scientific applications in her new role. “One of the exciting things about being a computer scientist at Berkeley Lab is that you are working alongside people who need help optimizing their scientific applications, and I get to bridge that gap between computer science and scientific discovery,” she said. “Data structures underlie everything in computer science. If you’re developing an application, you have a bunch of elements you need to store; I work on effectively storing those elements so that you can compute on them quickly.”
Xu notes that her work so far has been focused on building data structures for applications running on traditional single-node computers, so she’s excited to learn more about optimizing scientific codes to run on non-traditional heterogeneous supercomputers like NERSC’s Perlmutter system, which is comprised of CPUs and GPUs.
Although Xu’s Hopper Fellowship just began this month, she’s been working in the Lab’s Performance and Algorithms Research Group since February 2022, joining shortly after receiving her Ph.D. in Computer Science from the Massachusetts Institute of Technology (MIT).
She notes that her passion for computer science ignited when she was an undergraduate student at Stony Brook University.
“In high school, I was really into debating, theater, and music. It wasn’t until I got to college that I thought I’d try computer science. One of my first projects was to create an efficient version of a data structure called a skip list, a probabilistic data structure built on the idea of a linked list, and it was just so elegant,” said Xu. “I fell in love with computer science.”
She spent the next four years working with her advisor Professor Michael Bender on theoretical data structures for sparse graphs. When she got to MIT, her advisor, Professor Charles Leiserson, encouraged her to explore both the theory and the application and implementation of data structures.
“I took his course, and it was a steep learning curve. I spent most of my undergraduate years working on theory and doing a lot of math, but after this course, I became interested in applying data structures – how you make these things work in practice,” said Xu. She credits this course with sparking her interest in research opportunities at the national laboratories. Before coming to Berkeley Lab, Xu also spent some time interning at Sandia National Laboratory in Livermore, California.
A native of New York, Xu notes that she’s looking forward to the relatively warm winters on the West Coast. An avid hiker, she’s also excited to explore hiking trails around the Bay Area.
About Computing Sciences at Berkeley Lab
High performance computing plays a critical role in scientific discovery. Researchers increasingly rely on advances in computer science, mathematics, computational science, data science, and large-scale computing and networking to increase our understanding of ourselves, our planet, and our universe. Berkeley Lab's Computing Sciences Area researches, develops, and deploys new foundations, tools, and technologies to meet these needs and to advance research across a broad range of scientific disciplines.