Interactive Visualizations of Stellarscape

When you think of High Performance Computing you might not think about poetry, music, and dance. This past year, UA Dance, The Fred Fox School of Music, the School of Information, and the Department of Astronomy collaborated with High Performance Computing’s (HPC) visualization team on an immersive multimedia performance that synthesizes dance, music, science, visual art and technology.

The production, Stellarscape, tells the story of a massive star from birth to death through a combination of music, dance, and cinematography, blended with state-of-the-art data visualization and astrophysical simulation. Stellarscape premiered in January 2022 at UArizona and again at the installation of Wonderhouse at SXSW in Austin, Texas, in March.

The HPC visualization team supported the artists in scientific visualization and real-time interactive visual needs. The team made connections with established researchers and received permission to use simulation data and fixed media renders for the project. In addition, HPC resources were used to generate simulation snapshots from AGORA (Assembling Galaxies Of Resolved Anatomy) comparison projects and their initial condition datasets.

The Stellarscape team had access to a number of sensors from the new Health Sciences Sensor Lab, in order to track the dancers as they moved on stage. Part of their production budget was used to purchase a commercial license for TouchDesigner to incorporate data in interactive visualizations. This program made it possible to render the AGORA snapshots in real time as well as input from two different types of sensors.

A recording of the performance can be viewed at stellarscape.org.

Learn more about the HPC at rc.arizona.edu

Preparing for Additional Security Requirements

UITS supports Controlled Unclassified Information (CUI) research projects that require sensitive data handling to comply with federal government requirements. In FY22, the federal government announced that entities like the University of Arizona will be required to comply with additional security standards by 2025 to maintain annual research expenditures from the Department of Defense.

Research Technologies has partnered with the Information Security Office and University Export Control Office to plan for and implement a new Cybersecurity Maturity Model Certification (CMMC) 2.0 program. As part of the process, a CMMC self-audit was conducted to assess the University’s readiness. Several findings were discovered and remediation efforts are underway. The next step is an external pre-audit in 2023 to prepare for the final approval process.

Achieving the CMMC 2.0 certification is paramount to maintaining and, more importantly, growing the University’s research expenditures in fields where CUI is prevalent.

Breed Differences in Dog Cognition

Why do dog behaviorists need supercomputers? If you are trying to understand what dogs think and why, and you want to find the extent to which their behavior can be attributed to genetic traits, a supercomputer can be most helpful in running the thousands of analyses against a dataset of dog genetics.

Gita Gnanadesikan, a PhD candidate at the Arizona Canine Cognition Center, recently used HPC resources to analyze datasets for her scholarly article Breed Differences in Dog Cognition Associated with Brain-Expressed Genes and Neurological Functions. The article details the use of big genomic data to examine whether there are genetically based breed differences on cognitive and behavioral traits.

The cognitive dataset was collected through Dognition.com, a citizen science website that guides dog owners through a series of collective experiments. The genetic data was obtained from a publicly available data set that included 150,067 single nucleotide polymorphisms (SNPs) or genomic variants from over 1,400 dogs representing 161 breeds.

Gnanadesikan’s analysis used the applications Plink and R, both of which are available on HPC. One of the benefits of using HPC’s no-fee processing power is using it for analysis. HPC allows for thousands of complex analytical workflows, saving time from running these on a local workstation.