Central to the new approach is the concept of community software ecosystems. To work toward a sustainable path for next-generation computational science, the authors argue that community software ecosystem perspectives are needed — recognizing the vital role of software, not as just a byproduct of research on new algorithms or domain-specific science, but as an essential “first-class citizen.” The authors advocate that all member of the computational science community can play important roles in catalyzing change in projects, institutions and communities to address the technical, cognitive and social challenges of scientific software.
The authors lead ECP focus areas: Software Technology, Application Development, Hardware and Integration. The full paper presenting their discussion is available on the web:
Lois Curfman McInnes, Michael A. Heroux, Erik W. Draeger, Andrew Siegel, Susan Coghlan, and Katie Antypas, “How Community Software Ecosystems Can Unlock the Potential of Exascale Computing Systems,” Nature Computational Science, Feb. 22, 2021.