Production of an exascale-capable environment requires access to HPC resources for the purpose of application and software testing and development. Facility Resource Utilization is a collaboration between ECP and DOE HPC facilities to ensure effective access to current production systems and pre-exascale, and exascale computers.
For applications to take full advantage of exascale hardware and software, a robust developer training and productivity program keeps application and software team members, staff, and other stakeholders abreast of emerging technologies. This effort is a close collaboration of Facilities, vendors, and the ECP community.
Application Integration ensures that ECP applications are appropriately targeting DOE exascale architectures and that the facilities are aware of ongoing application development effort and can contribute their experience and expertise in porting applications to the facility architectures.
Through close partnership with ECP code teams, DOE HPC Facilities, and vendors, the Software Deployment team deploys and integrates an exascale software stack and deploys a software integration and testing capability at the Facilities to support continuous integration with site environments, including container technologies and software development kits.
The Message Passing Interface (MPI) is a community standard for inter-process communication and is used by the majority of DOE’s parallel scientific applications running on pre-exascale systems.
Led by the US Department of Energy’s (DOE’s) Lawrence Berkeley National Laboratory, the ExaStar project is working to deliver an exascale application for stellar explosions, including supernovae and neutron star mergers, as part of the DOE’s Exascale Computing Project (ECP). Such simulations will help scientists explain the origin of elements heavier than iron in the universe and answer other fundamental physics questions, such as the source of gravitational waves and the behavior of matter at extreme densities.
Small modular reactors (SMRs) and advanced reactor concepts (ARCs) will deliver clean, flexible, reliable, and affordable electricity while avoiding the traditional limitations of large nuclear reactor designs, including high capital costs and long construction timelines
The QMCPACK project is employing QMC methods to attack this problem since they robustly deliver highly accurate calculations of complex materials that do not artificially bias solutions of a given character.
The Adaptable I/O Systems (ADIOS) is designed to tackle data management challenges posed by large-scale science applications running on high-performance computers that require, for example, code-to-code coupling for multiphysics and multiscale applications and code-to-service coupling for data analysis and visualization.
Programmer productivity and performance portability are two of the most important challenges facing users of exascale architectures that include heterogeneous compute nodes, deep memory hierarchies, and persistent memory.