This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| ECP (Exascale Computing Project) | |
|---|---|
| Name | Exascale Computing Project |
| Abbreviation | ECP |
| Established | 2016 |
| Location | United States |
ECP (Exascale Computing Project). The Exascale Computing Project coordinates large-scale efforts to deploy exascale-capable computing systems in the United States, bringing together national laboratories, universities, and industry partners. It aligns research in hardware, system software, and scientific applications to accelerate capabilities for national priorities in science, energy, defense, and industry.
The project integrates work across Argonne National Laboratory, Oak Ridge National Laboratory, Lawrence Livermore National Laboratory, Sandia National Laboratories, Los Alamos National Laboratory, Pacific Northwest National Laboratory, National Energy Research Scientific Computing Center, Savannah River National Laboratory, Fermi National Accelerator Laboratory, Brookhaven National Laboratory to design exascale systems. It engages technology firms such as Intel Corporation, NVIDIA Corporation, AMD, IBM, Cray Inc., Hewlett Packard Enterprise, Google, Microsoft, Amazon Web Services and partners with academic institutions like Massachusetts Institute of Technology, Stanford University, University of California, Berkeley, University of Illinois Urbana–Champaign, Georgia Institute of Technology. The initiative interacts with agencies including the United States Department of Energy, National Nuclear Security Administration, Office of Science (United States Department of Energy), and coordinates with international research centers such as CERN and collaborations with projects like Top500 and HPCwire coverage.
Announced in 2016, the project was formed amid strategic guidance from leaders at Office of Science (United States Department of Energy), National Nuclear Security Administration, and directives influenced by reports from the National Academies of Sciences, Engineering, and Medicine, and recommendations tied to initiatives such as the National Strategic Computing Initiative. Early governance involved laboratory directors from Argonne National Laboratory, Oak Ridge National Laboratory, and Lawrence Livermore National Laboratory with program management coordinating via entities like the DOE Office of Technology Transitions and advisory input from panels including members affiliated with Stanford University, Princeton University, California Institute of Technology, University of Michigan, Columbia University. Organizational units combined efforts in application development, software technology, hardware and integration, and testbeds, while milestones referenced community benchmarks like Linpack and evaluations by groups including IEEE and ACM.
Primary objectives include delivering capable exascale systems that meet requirements for codes used in projects such as Nuclear Stockpile Stewardship Program and scientific campaigns in fusion energy research at facilities like ITER and climate modeling relevant to research by NASA and NOAA. Targets encompassed performance, energy efficiency, resilience, and programmability to support workflows from computational fluid dynamics in aerospace labs like NASA Ames Research Center to multiphysics simulations used by Sandia National Laboratories and Los Alamos National Laboratory. Objectives also addressed workforce development via partnerships with universities including University of Texas at Austin and University of California, San Diego and incorporation of standards from organizations like OpenMP, MPI Forum, and Khronos Group.
The project funded application teams focused on domains spanning materials science, climate modeling, nuclear physics, and machine learning, collaborating with researchers at Princeton Plasma Physics Laboratory, National Renewable Energy Laboratory, Argonne Leadership Computing Facility, and centers such as NERSC and ALCF. Software efforts emphasized portability and performance using ecosystems tied to OpenACC, CUDA, ROCm, libraries from MathWorks, and toolchains involving compilers from GNU Project, Intel Corporation, and LLVM Project. Application projects included work on codes that intersected with studies led by Lawrence Berkeley National Laboratory, University of Chicago, Columbia University and leveraged community projects like LAMMPS, GROMACS, S3D, and frameworks influenced by TensorFlow and PyTorch research. Development practices adopted provenance and reproducibility standards observed by repositories at GitHub and collaborative platforms used by teams at Carnegie Mellon University.
Hardware initiatives coordinated procurement and integration activities referencing architectures from Intel Corporation’s Xeon, AMD’s EPYC, accelerators from NVIDIA Corporation’s Tesla and AMD’s Radeon Instinct, interconnect technologies such as InfiniBand from Mellanox Technologies and novel memory technologies showcased by Micron Technology and SK Hynix. System integration involved national lab facilities like Oak Ridge Leadership Computing Facility and testbeds comparable to systems evaluated in TOP500 lists, with validation through benchmarks recognized by SPEC. Collaboration with vendors including Hewlett Packard Enterprise and legacy firms like Cray Inc. addressed thermal and power design challenges similar to studies at Lawrence Berkeley National Laboratory and design practices referenced in cases at Los Alamos National Laboratory.
Funding sources included programmatic allocations from the United States Department of Energy and investments coordinated with the National Nuclear Security Administration, supplemented by in-kind contributions from corporations such as Intel Corporation, NVIDIA Corporation, AMD, Hewlett Packard Enterprise and academic grants awarded through university partners including Massachusetts Institute of Technology and University of Illinois Urbana–Champaign. International collaborations and data-sharing agreements referenced engagement with institutions like CERN, European Centre for Medium-Range Weather Forecasts, and bilateral exchanges with research bodies such as Japan Aerospace Exploration Agency and French Alternative Energies and Atomic Energy Commission.
Achievements included preparation of production-ready application suites, advancements in scalable software libraries used by consortia including ACM and IEEE, and contributions to high-impact scientific results in domains connected to climate science research at NOAA and NASA, fusion modeling relevant to ITER, and materials design efforts linked to Argonne National Laboratory and Brookhaven National Laboratory. The project influenced procurement and architectural choices for exascale systems deployed at facilities such as Oak Ridge National Laboratory and Argonne National Laboratory and spawned workforce initiatives in collaboration with universities like Georgia Institute of Technology and University of California, Berkeley. Many technologies and methodologies developed under the project were reported in venues such as Supercomputing conference proceedings and journals associated with ACM and IEEE.
Category:High-performance computing projects