Ethan Kerzner, a senior mathematics and computer science double major from Edina, Minn., recently won a prize for a paper he wrote.
His paper, titled "Graphical Processing Unit Programming for Mathematical and Scientific Computing," won a Best Student Paper prize at the Midwest Instruction and Computing Symposium. The symposium was held at the University of Wisconsin-Eau Claire on April 16-17.
As part of the prize, Ethan received a monetary award provided by the sponsors of the conference, Digi-Key Corporation and CUNA Mutual Group.
His paper focused on the use of graphical processing units for mathematical and scientific computing, in particular the technology of NIVIDIA's Compute Unified Device Architecture (CUDA). Kerzner discussed an application of CUDA for the analysis of Hidden Markov Models and the errors in arithmetic of such an analysis.
Timothy Urness (left) and Ethan Kerzner (right)
Kerzner also addressed two possible solutions to the limited range of
the floating-point arithmetic. His adviser and co-author Timothy
Urness, Drake assistant professor of computer science, guided him
through the research.
Jason Grout, Drake assistant professor of mathematics, helped with the
project by putting Kerzner in touch with Glenn Tarbox, who works in
financial mathematics in Seattle. Tarbox agreed to let Kerzner use his
cutting-edge NVIDIA Telsa super-computer.
"This was done remotely, " Kerzner said. "I was able to log into his
computer to edit and run my programs. I used computers owned by Drake,
as well as Dr. Tarbox’s machine for my project. "
Kerzner will continue working with NVIDIA graphical processing units in
a summer research program at the University of Iowa. He was selected
for the program because of his experience in learning about CUDA with