ExaGraph: Combinatorial Methods for Enabling Exascale Applications
As a co-design center, ExaGraph will focus on developing combinatorial graph algorithms for several exascale application domains, including power grid, computational chemistry and biology, and Earth systems. The goal is to augment how high-volume data analytics are performed for applications and scientific computing. PNNL leads the project with contributors from Purdue University, and Lawrence Berkeley and Sandia national laboratories.
ExaLearn: Co-design Center for Exascale Machine Learning Technologies
The overarching goal of the ExaLearn co-design project is to provide exascale machine learning software for use by Exascale Computing Project (ECP) applications, other ECP co-design centers, and the U.S. Department of Energy (DOE) experimental facilities and leadership class computing facilities.