This Project Grant award from the National Science Foundation (NSF) Integrative Activities (IA) program in the amount of $271,883 supports research to improve software for graph computations through the development of a Graph Structure Descriptor Language and associated compiler infrastructure. The award to the University of Oklahoma aims to address the performance challenges of sparse and irregular computations, such as those involved in processing large-scale graph data, by developing improved tooling and integrating it into mainstream machine learning and graph analysis packages. This work seeks to translate recent performance gains in dense and regular computations to the sparse and irregular applications that are critical for understanding complex phenomena in fields like social network analysis, fraud detection, and disease modeling. The project is scheduled to run from January 1, 2025 to December 31, 2029.
Generated 2/25/25, 1:53 AM