UVM supercomputer to study climate, ‘online misinformation’

Addition of new data cluster could help prevent ‘spread of online misinformation,’ UVM says

UVM president Suresh Garimella speaks with Mike Austin, director of systems architecture, inside the Vermont Advanced Computing Core earlier this year. UVM photo

The University of Vermont’s supercomputer array will gain a massive new database cluster, dubbed DataMountain, that will allow researchers to “find solutions that improve the health of societies and the environment,” UVM said Sept. 27.

DataMountain will allow for near real-time access to enormous data files, supporting projects that require such speed to effectively analyze, describe, and explain rapidly growing datasets. The cluster will be available to the more than 500 Vermont researchers who have access to the university’s Vermont Advanced Computing Core (VACC) high performance computing environment.

With so many fields transitioning from data-scarce to data-rich environments, much important work will benefit from this new machine, including research into addiction, mental illness, climate change, drug discovery, food systems, and the spread of online misinformation. The cluster will support the work of UVM faculty from across the university, led by VACC director and principal investigator Chris Danforth (Mathematics & Statistics) and co-principal investigators Jianing Li (Chemistry), Meredith Niles (Food Sciences), Jarlath O’Neil-Dunne (Spatial Analysis Lab), Hugh Garavan (Psychiatry); and senior personnel Nick Cheney (Computer Science), and Peter Sheridan Dodds (Computer Science).

The National Science Foundation’s Major Research Instrumentation Program (MRI) and  Established Program to Stimulate Competitive Research (EPSCoR) have jointly awarded the university a new $1 million grant to build this large-memory machine. 

The new capability will allow for interactive access to over 50 terabytes of social media data for timely analysis of changes related to the COVID-19 pandemic, including physical and mental health data. It will also be used to help quantify the impacts of the COVID-19 pandemic on food insecurity, and explore massive amounts of genetic, task performance, and survey data associated with 10,000 adolescents across the United States over the next decade. 

The new database cluster will include 64 terabytes of random-access memory (RAM), spread across several large-scale servers connected by high-speed networking, and will create an extremely fast in-memory database system.

DataMountain will increase by nearly two orders of magnitude the largest random access memory machine currently available for computational research at UVM. It will also enhance the functionality of the high-performance computing clusters that make up the VACC facility—BlueMoon, DeepGreen, and BlackDiamond.

Funding for DataMountain arrives on the heels of a $1 million infrastructure upgrade to the VACC during fall 2021, and a gift of hardware support from corporate partner AMD in Fall 2020.

Categories: Education

1 reply »

Leave a Reply