cuDecomp
0.5.0
  • Overview
    • What is cuDecomp?
    • Decomposition Layout
  • Basic Usage Guide
    • Starting up cuDecomp
    • Creating a grid descriptor
    • Allocate pencil memory
    • Working with pencil data
      • C/C++
      • Fortran
    • Allocating workspace
    • Transposing the data
    • Updating halo regions
    • Cleaning up and finalizing the library
    • Building and running the example
  • Autotuning
    • Autotuning process
    • Autotuning usage
      • Creating a grid descriptor with autotuning enabled
      • Autotuner output and querying results
  • Working with NVSHMEM-enabled builds
    • Controlling the symmetric heap size
    • MPI compatibility
  • cuDecomp API
    • cuDecomp C API
      • Types
        • Internal types
        • Grid Descriptor Configuration
        • Pencil Information
        • Communication Backends
        • Additional Enumerators
      • Functions
        • Library Initialization/Finalization
        • Grid Descriptor Management
        • Workspace Management
        • Helper Functions
        • Transposition Functions
        • Halo Exchange Functions
    • cuDecomp Fortran API
      • Types
        • Internal types
        • Grid Descriptor Configuration
        • Pencil Information
        • Communication Backends
        • Additional Enumerators
      • Functions
        • Library Initialization/Finalization
        • Grid Descriptor Management
        • Workspace Management
        • Helper Functions
        • Transposition Functions
        • Halo Exchange Functions
  • Environment Variables
    • CUDECOMP_ENABLE_NCCL_UBR
    • CUDECOMP_ENABLE_CUMEM
cuDecomp
  • Search

© Copyright 2022, NVIDIA Corporation.

Built with Sphinx using a theme provided by Read the Docs.