Changes between Version 9 and Version 10 of Public/User_Guide/Offloading_hybrid_apps
- Timestamp:
- Sep 17, 2019, 3:24:04 PM (5 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
Public/User_Guide/Offloading_hybrid_apps
v9 v10 1 = '''Offloading computational tasks of hybrid MPI + OpenMP/OmpSs-2 applications to GPUs''' =1 = '''Offloading Computational Tasks of Hybrid MPI + OpenMP/OmpSs-2 Applications to GPUs''' = 2 2 3 3 Table of contents: … … 34 34 35 35 * '''GNU Compiler Collection'''. 36 * '''!OmpSs-2''': !OmpSs-2 is the second generation of the '''!OmpSs''' programming model. It is a task-based 36 * An '''MPI''' implementation supporting the multi-threading level of thread support. 37 * The '''Task-Aware MPI (TAMPI)''' library which provides a clean interoperability mechanism 38 for MPI and OpenMP/!OmpSs-2 tasks. Downloads and info at [https://github.com/bsc-pm/tampi]. 39 * The '''!OmpSs-2''' model which is the second generation of the '''!OmpSs''' programming model. It is a task-based 37 40 programming model originated from the ideas of the OpenMP and !StarSs programming models. The 38 41 specification and user-guide are available at [https://pm.bsc.es/ompss-2-docs/spec/] and … … 40 43 '''Nanos6''' tools. Mercurium is a source-to-source compiler which provides the necessary support for 41 44 transforming the high-level directives into a parallelized version of the application. The Nanos6 42 runtime system libraryprovides the services to manage all the parallelism in the application45 runtime system provides the services to manage all the parallelism in the application 43 46 (e.g., task creation, synchronization, scheduling, etc). Downloads at [https://github.com/bsc-pm]. 44 * '''Clang + LLVM OpenMP''' (derived): 45 * '''MPI''': This application requires an MPI library supporting the multi-threading level of 46 thread support. 47 48 In addition, there are some optional tools which enable the building of other application versions: 49 50 * '''CUDA''' and NVIDIA '''Unified Memory''' devices: This application has CUDA variants in which some of 47 * A derivative '''Clang + LLVM OpenMP''' that supports the non-blocking mode of TAMPI. 48 * The '''CUDA''' tools and NVIDIA '''Unified Memory''' devices for the CUDA variants, in which some of 51 49 the N-body kernels are executed on the available GPU devices. 52 * '''Task-Aware MPI (TAMPI)''': The Task-Aware MPI library provides the interoperability mechanism53 for MPI and OpenMP/!OmpSs-2. Downloads and more information at [https://github.com/bsc-pm/tampi].54 50 55 51 === Versions ===