Purdue University

A nuclear engineering graduate student at Purdue has partnered with RCAC staff to use the Negishi cluster to refine and scale his model.

Samuel Wyss, who recently graduated with his master’s degree in nuclear engineering under the supervision of Allen Garner, studies a phenomenon known as electroporation, where cell membranes open up in response to a high intensity electromagnetic field. This has important applications to healthcare, allowing for more efficient drug delivery and better cancer treatments.

While electroporation models exist in the literature, Wyss and his team are the first ones to parallelize them to model cellular level effects.

Wyss sought out RCAC’s coffee hour consultations, where expert staff offer support to Purdue researchers who are using or interested in using high-performance computing in their work. There, he connected with Geoffrey Lentner, lead research data scientist for RCAC, who helped Wyss refine his model, and offered a loan of Negishi nodes to scale the model.

“I’ve learned so much from Geoffrey and the other RCAC staff members and they’ve been a great help,” says Wyss. “These are things that you can’t learn in class. There are classes that teach the basics of high-performance computing, but through my discussions with RCAC I’ve learned a lot more about the application side of it, which I’ve really enjoyed.”

Running the models on a personal computer, it would take days to model a single set of parameters. Wyss had the idea to parallelize the models on Negishi to accelerate the process and run the model for many different parameters, quickly getting more information that can be used by experimentalists doing studies on real cells.

“Without the Negishi cluster, this work wouldn’t have been possible,” he says.

To learn more about Negishi and other RCAC resources, contact rcac-help@purdue.edu and visit the RCAC website.