Heterogeneous catalysis and the design of new catalysts is a grand challenge problem that will require the availability of exascale computers. In order to take full advantage of exascale architectures, it is critical that application software be developed that is capable of scaling to millions of cores and take advantage of emerging low power architectures that dramatically lower the energy/power cost without significant deterioration of time to solution.
This ECP effort will develop ab initio methods in the electronic structure program GAMESS based on fragmentation methods that already scale beyond the petascale combined with quantum Monte Carlo. In order to attain the exascale, GAMESS will be refactored to take advantage of modern computer hardware and software, and the capabilities of the C++ libcchem code that is co-developed with GAMESS will be greatly expanded. Concurrently, performance analyses will be done for the broad array of electronic structure methods in GAMESS on current and emerging architectures (e.g., ARM, GPU, Phi) to assess their ability to increase time to solution while decreasing energy demands. The new codes that are developed will be brought to bear on the heterogeneous catalysis problem, specifically using mesoporous silica nanoparticles, including tens of thousands of atoms, as a template.
The PIs for this ECP effort comprise a mix of computer scientists and application scientists who have a successful, productive history of collaborations. The easily downloadable new codes will be shared with the broad community at no cost. Close collaborations are anticipated with the NWChem and QMCPACK ECPs.