Wide-scale Adoption through Visually-Enhanced HPC codes (WAVE-HPC) for semiconductor R&D

While many ASCR-supported software packages are open source, these software packages are complicated to use, posing a significant barrier to many non-expert organizations. Even when the expertise required to install and use these packages is present, potential adopters lack options for the level of support, training, and personalized development required for usage across the organization. Without a commercial interest in marketing the capabilities of this software, broader adoption is limited by a lack of availability within the wider industrial and technological ecosystem. Exabyte Inc. proposes to develop an open-source addition to its platform that provides a way to package multiscale modeling and simulation tools based on the Extreme-scale Scientific Software Stack (E4S), and deploy them to accelerate novel materials R&D through the optimized use of computing resources for key tasks. The tools will prioritize ease of deployment for industrial users and leverage Exabyte’s existing platform and the direct input of existing semiconductor materials customers to inform and validate design choices. During Phase I, we propose to install E4S libraries to (1) conduct partner research on the desired design, implementation, and usability of tools for incorporation; (2) develop data standards for the corresponding workflows and materials properties to simplify user input and decrease complexity; (3) work with our industrial partner to install and test these capabilities. This proposal builds upon the ideas covered by the Materials Genome Initiative [MGI]. It is designed to help disseminate the DOE HPC code development efforts to a broader cross-section of US materials scientists. Our initial focus on the semiconductor market is guided by both the projected growth of as high as 50% and due to its projected outsized impact on global energy demand – accounting for some 20% of the total by 2030.