NCSA Groups Sweep Original Undergraduate Research Awards at EOH 19

Three undergraduate student groups from the National Center for Supercomputing Applications (NCSA) have been awarded the top three places in this year’s Illinois Engineering Open House (EOH) Original Undergraduate Research awards. The awards, which were chosen by The Grainger College of Engineering alumni judges, were selected out of the entire Illinois engineering campus and took into account the research itself, the ease of explanation of the research, and the exhibitors ability to engage with the EOH Audience.

First place went to the Student Cluster Competition team, capping off what has been a brilliant year thus far. At SC18 in Dallas in November, they took home first place out of all American competitors, their best-ever finish at the annual Supercomputing Conference.

The team, which is made up of entirely undergraduate students, seeks to both construct an HPC cluster and then have it perform a series of rigorous tasks within time and power-usage constraints. Yan Zhang, team member in 2016 and 2017, and now Assistant System Engineer at NCSA, presented the system used by the 2018 team at EOH.

“This is purely a student-driven effort that lets students experience working with high-performance computer systems typically found at supercomputing centers and typically out of reach of undergraduate students,” said team advisor and NCSA Senior Research Scientist Volodymyr Kindratenko.

The second place award for Original Undergraduate Research went to the NCSA Genomics team, who represented multidisciplinary work done by the NCSA Industry program on the Blue Waters and iForge supercomputers at NCSA to model challenging biomechanical problems and to analyze massive data sets for advanced healthcare.

“With experts from multiple disciplines, NCSA Industry successfully executed highly collaborative cutting-edge research that has addressed disparate challenges in healthcare,” said NCSA Graduate Hourly Project Manager Priya Balgi.

“The noteworthy research presented at the EOH in the field of Bio-Multiphysics and BioInformatics was the Alya Cardiac Computational Model and the Mayo Grand Challenge Project. The Alya Cardiac Computational Model, a product of a collaboration with the Barcelona Supercomputing Center (BSC), was made possible by scaling Ayla code to 100,000 processors on the Blue Waters supercomputer, where the execution could be completed in less than two hours instead of approximately 18 years. The Mayo Grand Challenge project developed a genomic workflow so that robust analyses could be performed on complete genome sequences of hospital patients.”

NCSA’s Gravity Group also took home third place in the EOH Original Undergraduate Research award this year, their second in a row. The group, which has a focus on Deep Learning and Machine Learning, investigates the physics of the universe, gaining insights into astrophysical phenomena with the help of advanced computing, like the Blue Waters system.

The NCSA Gravity group has used these techniques to analyze everything from black hole collisions to gravitational waves, and even constructing catalogs of galaxies.

“At the interface of this work, they are designing algorithms to train neural networks with supercomputers, facilitating the use of deeper models that can be trained with ever-larger data sets,” mentioned Eliu Huerta, NCSA Gravity Group Lead, “It is a pleasure to work with them.”

In addition to these awards, NCSA put on 15 of exhibits this year, spanning a diverse array of disciplines and showcasing a variety of the groundbreaking research being done here, and at the University of Illinois as a whole.

NCSA would like to sincerely thank all of the students and staff who made the exhibits and demonstrations at Engineering Open House 2019 such a success.

Disclaimer: Due to changes in website systems, we've adjusted archived content to fit the present-day site and the articles will not appear in their original published format. Formatting, header information, photographs and other illustrations are not available in archived articles.