A Florida State University researcher is leading a $4.4 million Department of Energy project to help create software that can take advantage of supercomputer capabilities and advance quantum information science.
The project is led by FSU Associate Professor of Chemistry and Biochemistry Eugene DePrince and includes collaborators from Virginia Tech, University of Washington, and Lawrence Berkeley National Laboratory.
"This is a major scientific software engineering project, and we want our software to be deployable on tens of thousands of computer nodes that might include both conventional CPUs and accelerators like graphical processing units (GPUs)," DePrince said. "Our team includes experts in accurate quantum mechanical theories for electronic systems, the dynamics of electronic systems, relativistic theories and high-performance computing."
Quantum science has become a major priority for the U.S. Department of Energy. DePrince's grant is part of an overall $28 million investment by DOE in five research projects to develop software that will allow supercomputers to make giant leaps in quantum information science.
Quantum mechanics is a fundamental physics theory that allows for the calculation of properties and behaviors of microscopic systems such as molecules, atoms and sub-atomic particles. Computers, lasers and scanning technologies are all based off quantum mechanical systems.
DePrince’s initial work will focus on developing capabilities to simulate the dynamics of quantum-mechanical systems that could be used to create spin-based technologies. Though he will be working on microscopic systems, DePrince noted that similar simulations could be applied to everyday objects — a baseball for example — although the math is a little different.
"Everyone has an intuitive feeling for this, even if they aren't familiar with the math," he said. "If you throw a ball, you can probably guess how far it is going to go."
DePrince and his colleagues will use the laws of quantum mechanics to calculate how different quantum mechanical systems behave over time. They want to be able to calculate the wave function — a mathematical function that describes the location of an electron at a given point in space — so that they can make predictions about the properties of the system at that time.
This can be further complicated if the systems they’re modeling involve the heavier chemical elements that are found at the bottom of the periodic table, such as uranium or plutonium, because they require the consideration of relativistic effects.
“The mathematics involved can get pretty complicated,” DePrince said.
The DOE program funding the project is called the SciDAC Partnership or Scientific Discovery through Advanced Computing, and these partnerships are meant to fund collaborations between domain experts and DOE computational scientists in one of their SciDAC institutes.
The collaborating DOE scientist for this project is Chao Yang from Lawrence Berkeley National Laboratory.
This funding will also support two postdoctoral scholars and part of a graduate researcher’s salary for four years.