To produce the image of a dying star, Evan Schneider tapped into “El Gato.”

Evan Schneider

This is a computer-generated simulation of what happens when the explosion of a dying star plows through a cold dust cloud that is surrounded by lower density gas.

The supernova shock condenses and heats the gas in the cloud, producing the hot regions of yellow and red in the image. It is a microcosm of the cosmic environment in which stars are born, die and explode to affect the structure and evolution of galaxies and the formation of solar systems.

Theoretical astrophysicists study these simulations to learn about how our cosmos is formed.

This simulation is one of the first products from a software code created by Evan Schneider, a graduate student in astrophysics at the University of Arizona.

Schneider recently won the 2014 UA Theoretical Astrophysics Program student research prize for the creation of “Cholla” — a massively parallel hydrodynamics code for astrophysical simulation.

It runs 50 times faster than a currently used program by harnessing the power of a bank of 140 graphics processing units linked to a powerful conventional computer to create the UA’s “El Gato” supercomputer.

Schneider said in her TAP Prize talk that she is anxious to put her code to work.

“Creating code is great, but I’ve been doing it for two years and science is why we come to the University of Arizona for grad school,” she said.

Graphics-processing units, or GPUs, were “developed to render pixels on a computer screen very quickly,” she said. Taking that ability from video games and smartphones to astrophysical simulation is a relatively recent development, said Schneider.

UA Regents’ Professor Dave Arnett, who pioneered the use of supercomputers in astrophysical research, watched Schneider’s presentation and told her he had already thought of several ways to employ her code in research.

“We’ve gone from molasses to warm molasses and now this. This is exciting,” said Arnett.

Schneider said the program offers a way to save both time and energy. “For the people concerned about the environment, it’s energy-efficient and for those concerned about money — it’s cheap.”

The simulation depicted here is “really quite tiny on the scale of a whole galaxy,” Schneider said. “We want to have a simulation with a high-enough resolution where we can model the entire galaxy.”

Schneider’s adviser and co-author in the work is Steward astronomer Brant Robertson, who led the bid to bring the El Gato computer to the UA, winning a $1.3 million grant from the National Science Foundation’s Division of Astronomical Sciences.

Contact reporter Tom Beal at or 573-4158.