34 miljoen dollar voor softwareprojecten rond exascale computing

Het Exascale Computing Project (ECP) van het Amerikaanse Department of Energy stelt 34 miljoen dollar beschikbaar voor 35 softwareprojecten. De projecten die worden gefinancieerd zijn allen gericht op exascale computing, een vorm van computersystemen die tenminste 50 maal sneller zijn dan de meest krachtige supercomputers van dit moment.

Geïnteresseerden konden voorstellen indienen bij het ECP voor softwareprojecten die betrekking hebben op exascale computing, met het verzoek hiervoor subsidie te ontvangen. De 35 softwareprojecten die door de DoE zijn geselecteerd hebben betrekking op verschillende componenten van de software stack voor exascale systemen, waaronder programmeermodellen en runtime libraries, raamwerken, tools, systeemsoftware, datamanagement, I/O, visualisatie en data analyse.

Hieronder een overzicht van de softwareprojecten die financiering ontvangen vanuit het ECP:

• xGA: Global Arrays on Extreme Scale Architectures - Abhinav Vishnu, Pacific Northwest National Laboratory (PNNL)
• Integrated Software Components for Managing Computation and Memory Interplay at Exascale - Richard Hornung, Lawrence Livermore National Laboratory (LLNL)
• Lightweight Communication and Global Address Space Support for Exascale Applications - Scott Baden, Lawrence Berkeley National Laboratory (LBNL)
• Exascale MPI, Pavan Balaji - Argonne National Laboratory (ANL)
• Enhancing and Hardening the Legion Programming System for the Exascale Computing Project - Galen Shipman, Los Alamos National Laboratory (LANL) with ANL, Stanford Univ.
• Distributed Tasking for Exascale - Jack Dongarra, Univ. of Tenn.
• ECP Applications effective use of Kokkos to achieve performance portability across exascale architectures - Carter Edwards, Sandia National Laboratories (SNL) with LANL, Oak Ridge National Laboratory (ORNL)
• Open MPI for Exascale (OMPI-X) - David Bernholdt, ORNL with LANL, LLNL, SNL, Univ. of Tenn.
• A Runtime System for Application-Level Power Steering on Exascale Systems - Martin Schulz, LLNL with Univ. of Arizona
• SOLLVE: Scaling OpenMP with LLVM for Exascale performance and portability - Barbara Chapman, Brookhaven National Laboratory (BNL) with ANL, LLNL, ORNL, Rice Univ., UIUC
• PROTEAS: PROgramming Toolchain for Emerging Architectures and Systems - Jeffrey Vetter, ORNL with ANL, LANL, Univ. of Oregon
• Extending HPCToolkit to Measure and Analyze Code Performance on Exascale Platforms - John Mellor-Crummey, Rice Univ. with Univ. of Wisconsin
• Autotuning Compiler Technology for Cross-Architecture Transformation and Code Generation - Mary Hall, Univ. of Utah with ANL, LBNL
• EXA-PAPI: The Exascale Performance Application Programming Interface - Jack Dongarra, Univ. of Tenn.
• Exascale Code Generation Toolkit - Dan Quinlan, LLNL with PNNL, Ohio State Univ., Colorado State Univ.
• Software for Linear Algebra Targeting at Exascale (SLATE) - Jack Dongarra, Univ. of Tenn.
• Production-ready, Exascale-Enabled Krylov Solvers (PEEKS) for Exascale Computing - Mike Heroux, SNL with Univ. of Tenn.
• ForTrilinos: Sustainable Production Fortran Interoperability with Trilinos Libraries - Mike Heroux, SNL with LANL, ORNL
• Factorization Based Sparse Solvers and Preconditioners for Exascale - Sherry Li, LBNL
• Extreme-scale Scientific Software Development Kit for the Exascale Computing Project: xSDK4ECP - Lois Curfman McInnes, ANL with LBNL, LLNL, SNL, Univ. of Tenn, UC Berkeley
• ALExa: Accelerated Libraries for Exascale - Wayne Joubert, ORNL
• Enabling Time Integrators for Exascale Through SUNDIALS - Carol Woodward, LLNL with Southern Methodist Univ.
• Preparing PETSc/TAO for Exascale - Barry Smith, ANL with Univ. of Colorado, Rice Univ.
• VeloC: Very Low Overhead transparent multilevel Checkpoint/restart - Franck Cappello, ANL with LLNL
• EZ: Fast, effective, parallel error-bounded exascale lossy compression for scientific data - Franck Cappello, ANL
• UNIFYCR: A Checkpoint/Restart File System for Distributed Burst Buffers - Kathryn Mohror, LLNL with ORNL
• ExaHDF5: Delivering Efficient Parallel I/O on Exascale Computing Systems - Suren Byna, LBNL with ANL, The HDF Group
• The ADIOS framework for Scientific Data on exascale systems - Scott Klasky, ORNL with LBNL, Kitware, Rutgers Univ.
• Data Libraries and Services Enabling Exascale Science - Rob Ross, ANL with LANL, Northwestern Univ.
• ZFP: Compressed Floating-Point Arrays - Peter Lindstrom, LLNL
• ECP ALPINE: Algorithms and Infrastructure for In Situ Visualization and Analysis - James Ahrens, LANL with LBNL, LLNL, Univ. of Oregon, Kitware
• ECP VTK-m: Updating HPC Visualization Software for Exascale-Era Processors - Ken Moreland, SNL with LANL, ORNL, Univ. of Oregon, Kitware
• Enhancing Qthreads for ECP Science and Energy Impact - Ron Brightwell, SNL
• 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