Lawrence Livermore scientists are among those awarded funding to develop software for the Department of Energy’s Exascale Computing Project (ECP). The ECP announced today the selection of 35 software development proposals representing 25 research and academic organizations.
The awards for the first year of funding total $34 million and cover many components of the software stack for exascale systems, including programming models and runtime libraries, mathematical libraries and frameworks, tools, lower-level system software, data management and I/O (input/output), as well as in situ visualization and data analysis.
Exascale refers to computing systems at least 50 times faster than the nation’s most powerful supercomputers in use today (systems capable of at least one exaflop, a billion billion calculations a second).
Livermore computer scientists will lead six of the projects and are collaborators on another seven. Project leaders include Richard Hornung, Martin Schulz, Dan Quinlan, Carol Woodward, Kathryn Mohror and Peter Lindstrom.
The list of selected proposals involving LLNL, principal investigators and collaborating organizations are as follows:
- Integrated Software Components for Managing Computation and Memory Interplay at Exascale, Richard Hornung, Lawrence Livermore National Laboratory (LLNL)
- Open MPI for Exascale (OMPI-X), David Bernholdt, Oakridge National Laboratory (ORNL) with Los Alamos National Laboratory (LANL), LLNL, Sandia National Laboratory (SNL) and University of Tennessee
- A Runtime System for Application-Level Power Steering on Exascale Systems, Martin Schulz, LLNL with University of Arizona
- SOLLVE: Scaling OpenMP with LLVM for Exascale Performance and Portability, Barbara Chapman, Brookhaven National Laboratory (BNL) with Argonne National Laboratory (ANL), LLNL, ORNL, Rice University, University of Illinois at Urbana-Champaign (UIUC)
- Exascale Code Generation Toolkit, Dan Quinlan, LLNL with Pacific Northwest National Laboratory (PNNL), Ohio State University, Colorado State University
- Extreme-scale Scientific Software Development Kit for the Exascale Computing Project: xSDK4ECP, Lois Curfman McInnes, ANL with Lawrence Berkeley National Laboratory (LBNL), LLNL, SNL, University of Tennessee, UC Berkeley
- Enabling Time Integrators for Exascale Through SUNDIALS, Carol Woodward, LLNL with Southern Methodist University
- VeloC: Very Low Overhead Transparent Multilevel Checkpoint/Restart, Franck Cappello, ANL with LLNL
- UNIFYCR: A Checkpoint/Restart File System for Distributed Burst Buffers, Kathryn Mohror, LLNL with ORNL
- ZFP: Compressed Floating-Point Arrays, Peter Lindstrom, LLNL
- ECP ALPINE: Algorithms and Infrastructure for In Situ Visualization and Analysis, Jim Ahrens, LANL with LBNL, LLNL, University of Oregon, Kitware
- A Simplified Complex Memory API and Operating System/Runtime Interface for ECP, Mike Lang, LANL with LLNL, ORNL, SNL, Georgia Tech
- Argo: Operating system and Resource Management for Exascale, Pete Beckman, ANL with LLNL
See the full press release.
johnston19 [at] llnl.gov