Skip to main content

Big Red II in Review: Five years of HPC-powered research

High performance systems May 8, 2020

2019 saw the retirement of Big Red II, once IU’s primary system for high performance parallel computing dedicated to research. In five years, it supported over 1,700 users in 237 academic disciplines, handling over 2.5 million jobs with nearly 9 million core hours delivered.

Here, we take a look back at some of the research impacted by Indiana University’s commitment to providing HPC resources.

Yun Liu with a 3D-printed model of the chloride-capture molecule.

Photo by Fred Zwicky, University of Illinois, Urbana-Champaign

Salt-extraction molecule discovered with predictive modeling: 

Yun Liu, a former PhD student in the lab of Amar Flood, led a team of researchers to discover a molecule that desalinates water and soil. The findings, published in Science, relied on molecular modeling. “To capture the real-world structure of the molecule as accurately as possible, we run extensive calculations taking into account the environment in which the molecule will be used, like the solution the molecule will dissolve in,” said Liu. Researchers say Big Red II reduced the time-to-science for this experiment, estimating the time needed for computational analysis dropped from months to a few days.

Using simulations to understand the universe’s building blocks:

Researchers at Indiana University and Caltech employed Big Red II to recreate the forces of space to understand the properties of nuclear pasta - the theoretical material found at the inner crust of neutron stars. Big Red II completed about two million hours of processor time, “doing the largest simulations ever of the densest material in the universe,” said researcher Matthew Caplan, who also lauded the UITS support team for its help behind the scenes. Publishing their observations in Physical Review Letters, the team discovered a great deal of pressure was necessary to bend nuclear pasta; the pressure required to break it, however, was greater than for any known material.

3d illustration of a ribonicleic acid chain from which the deoxyribonucleic acid or DNA is composed


Examining the power of “junk” RNA with statistical models: 

Felipe Porto, a Master’s student in Dr. Sarath Chandra Janga’s lab at IUPUI, used Big Red II to observe long non coding RNA or IncRNA. Long dismissed as “junk DNA,” Porto’s work views IncRNA sequences as a giant, jumbled up sentence. With experiments splicing the “grammar” of the genetic code within cancer cells, Porto observes changes in gene expression. The data storage and processing capacity of Big Red II made the research lab “avid users and big fans” of HPC resources, said Porto.

Big Red II by the numbers:

Total Jobs submitted: 2,546,563
Total Core Hours delivered: 891,965,342
Total Unique Job-running Users: 1,739
Total Disciplines: 237
Total Unique Binary Count: 7,950

Big Red 3, a Cray XC40 supercomputer, replaced Big Red II in early 2020. For more information on accessing Big Red 3 for research, check out IU’s Knowledge Base.

More stories