Aric Hagberg
Aric Hagberg’s analyses
of complex pattern formation have
gained an unusual fan: a Boston artist who bases works on
Hagberg’s colorful demonstrations.
Hagberg, a DOE CSGF fellow from 1992 to 1994, has a broad portfolio as a scientist at Los Alamos National Laboratory. The group he belongs to focuses on mathematical modeling and analysis, but Hagberg has dabbled in computer networking, programming, and even assembly of an early high-performance computer cluster.
Hagberg’s doctoral dissertation and subsequent research focus mainly on mathematical analysis of how chemical patterns propagate in systems like the Belousov-Zhabotinsky reaction. The mathematical models — and experiments they’re compared against — generate pictures and movies of spirals, vortices, and labyrinths. (Go to http://math.lanl.gov/~hagberg/Movies/)
That caught the attention of artist Brian Knep, who asked to borrow the computer codes for the simulations. “I thought, ‘Should I answer some crazy artist’s questions?’” Hagberg recalls. Hagberg sent Knep some information, and the artist took it from there. His computer creations interact with gallery visitors and mimic patterns found on Hagberg’s Los Alamos Web site. (See Knep’s work at http://www.blep.com/projects.htm)
“They’re beautiful. He really captured the essence,” Hagberg says. “He is a very skilled software engineer. I’ll send him anything he wants now.”
One example is Hagberg’s work on network dynamics. In one project, he’s working with biologists and engineers to improve wireless sensor networks for ecological research. The idea is to have sensors analyze data and make initial judgments about their quality. That will make it easier for biologists to sift the piles of information the sensors generate.
Hagberg also authored NetworkX, an open-source program to create, manipulate and analyze complex networks. People from around the world, including the U.S. military, have used it.
It’s typical for mathematical modeling and analysis researchers to dive into such a variety of projects, Hagberg says. “There’s 30 of us and there are probably 40 or 50 projects” they’re working on, he adds. “We’re like consultants. We search for interesting and hard problems.”
That spirit led Hagberg to collaborate with other Los Alamos researchers on assembling what was rated as the 113th fastest machine in the world in 1998. Spending just $313,000, the team bought 140 personal computers by mail order and connected them in a cluster. “We showed people you could do it. You could definitely build a Top 500 supercomputer” for less cost with commercial technology, Hagberg says. “Now, you go into any university, and they have lots of these Linux clusters around.”
The exercise illustrates the entrepreneurial spirit Hagberg enjoys at Los Alamos. He sounds like someone destined to be there a long time.
“I know a good thing when I see it,” Hagberg adds. “It’s a great place to work, and I was lucky to find it.”
