Introducing Puma High Performance Computing
The University of Arizona invests in computing infrastructure to empower the growing research community every 4 years. University research benefits from free computing resources in several ways; the most important is that new research faculty do not have to spend startup funds and critical time to set up servers to start their research. Major University projects and institutes also use Arizona’s high performance computing (HPC) such as OSIRIS-REx, TERRA-REF, Event Horizon Telescope, iMicrobe, the Arizona Cosmology Lab, and the Cancer Center Bioinformatics Shared Resource (BISR).
UITS Research Computing consultants are available to support researchers at any point in their work: experimental design, handling large-scale data, analysis on Puma, statistical analysis, and data visualization.
Puma was purchased after an open Request for Proposals (RFP) process directed by the Research Computing Governance Committee as well as UITS Research Technologies. Twelve vendors responded and their bids were compared based on a structured decision matrix. Puma was stress-tested on Folding@home to support COVID-19 research before it was released to the UA research community.
Three Arizona Universities Team Up to Identify COVID Protein Folding
Arizona’s three public universities participated in the national Folding@home project, which relies on volunteers’ idle computing power to run protein modeling computations that help researchers learn more about how to cure or treat certain diseases.
The project seeks to understand how proteins – large, complex molecules that play an important role in how our bodies operate – “fold” to perform their biological functions. This helps researchers understand diseases that result from protein “misfolding” and identify novel ways to develop new drug therapies. How proteins fold or misfold can help researchers understand what causes diseases like cancer, Alzheimer’s disease and diabetes. This past year’s focus was on providing insight into COVID-19.
Volunteers can track their contributions on the Folding@ home website and combine their efforts as a team, receiving points for completing work assigned to them. The point system helps Folding@home determine which machines in the project are quick and reliable.
As of early July 2020, the team was ranked in the top 53 out of nearly 250,000 teams, surpassing other teams that include Hewlett Packard, Cisco Systems, Apple Inc. and Google, as well as many other universities, industry and national or international contributors. When the University of Arizona’s PUMA came online, the team’s ranking quickly jumped to 37.
Memory Jogger Assists Campus Community in Tracing Their Location
When members of the campus community are notified they have tested positive, they may need help remembering where on campus they have been. The Memory Jogger program was developed as an anonymous way to help them jog their memory!
Memory Jogger was created to complement other COVID tracing activities on campus, such as the COVID Watch anonymous exposure notification application, for students, faculty, and staff who have been informed they have tested positive for COVID-19.
Campus Wi-Fi data and Splunk monitoring capability feeds into a campus mapping tool where the COVID positive individual can access their own anonymized location data on the University campus map. It’s all automated, and due to privacy considerations, the data report is generated unique to them so no one else can see the data.
UITS Research Participates in National Consortium
UITS Research Technologies was extended an invitation in 2019 to sit a staff member on the project evaluation committee for the COVID-19 HPC Consortium, which is a part of the cloud-based, on-demand computing and data analysis resources within the national Extreme Science and Engineering Discovery Environment, known as XSEDE. The consortium is a unique private- public effort spearheaded by the White House Office of Science and Technology Policy, the U.S. Department of Energy, and IBM to to bring together federal government, industry, and academic leaders who are volunteering free compute time and resources on their world-class machines.
Research Data Center Usage
Principal Investigators (PIs) using HPC systems
Active awards using HPC systems
Active Researchers Using HPC
Top 100 PIs using HPC
Total CPU Ocelote, El Gato, and Puma capacity
Total Cores of All HPC Systems
Faculty Compute Allocation
Yearly Faculty Compute Hours Allocation
- Supercomputing (HPC)
- Regulated Research Environment
- Research Support Services
- UA Vitae
Number of AMD EPYC 7642 Cores in Puma
(2X the size of Ocelote and 100X of El Gato)
Total expenditures by HPC investigators
Total sponsored research expenditures by investigators using HPC services
"The HPC has enabled research that I would not have been able to accomplish otherwise. I simply would not have been able to process all the data required for my regional photometry analysis in a reasonable amount of time."
Dathon Golish, PhD, Imaging Scientist, OSIRIS-REx Mission