Indiana University will receive $580,000 of a $3.2 million grant for its collaboration in a National Science Foundation Cyberinfrastructure for Science Innovation project. Dingwen Tao, associate professor of Intelligent Systems Engineering for the Luddy School of Informatics, Computing, and Engineering, is IU’s primary investigator.
The University of Chicago, Ohio State University and the University of Alabama also are collaborating on the project, which aims to develop sustainable computer software to support science research and discovery.
CSSI is the largest software development program across multiple NSF directorates.
"Professor Tao’s research will contribute significantly to the proposed data management ecosystem," said Beth Plale, ISE chair and Michael A and Laurie Burns McRobbie Bicentennial Professor of Computer Engineering. “The award also contributes to distinguishing the Intelligent Systems Engineering Department’s crucial role in innovation and research translation.”
Tao said the project will build a large ecosystem to support about 10 science domains for their efficient data management. It will more swiftly and efficiently transmit, store and analyze data, which will have dramatic impact in areas such as cosmology, climate, seismology, molecular dynamics, material science, combustion, quantum simulation, and plasma physics.
“Winning this award will enable us to continue our collaborative efforts after the Exascale Computing Project concludes, and will also increase the sustainability of the scientific software we've developed,” Tao said. “It also will provide educational and training opportunities for undergraduates, graduates, and cyberinfrastructure professionals.”
Simulations and experiments at scientific facilities generate more data than can be efficiently transmitted, stored, and analyzed. That makes data reduction a key part of scientific data analysis and conservation workflows. Lossy compression, a vital data reduction technique, has improved significantly in recent years. However, existing generic lossy compressors for scientific data often fall short of providing optimal performance for each application domain and use case. This has spurred development of specialized lossy compressors that outperform generic versions.
Designing and implementing specialized lossy compressors is complex. It requires expertise from domain scientists and specialists in lossy compression algorithms and hardware systems implementations. The process can take years.
This project will create a framework called “FZ” to enable users to compose and use novel, specialized lossy compressors more swiftly. The framework caters to their domains, applications, use cases, and requirements.
The Luddy team will collaborate with 10 science partners to leverage this framework and develop specialized lossy compressors for their science applications.
“The NSF has made numerous investments in scientific software development through large Software Institutes and other programs,” Tao said. “Our FZ infrastructure will significantly streamline the design and development process of new specialized lossy compressors, enhancing efficiency in scientific data reduction and management. “
Funding begins on Aug. 1 and is projected to end July 31, 2027.
Tao also will collaborate with the cyberinfrastructure team to develop IU’s cyberinfrastructure workforce and maintain the university’s position as a leading institution in cyberinfrastructure research and development.
Since 2016, this team has worked with the U.S. Department of Energy’s Exascale Computing Project, which includes members from multiple universities and national laboratories.
NSF is an independent federal agency that supports science and engineering in all 50 states and U.S. territories.