Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/sandialabs/p3a
Portably Performant Physical Algebra
https://github.com/sandialabs/p3a
amd-gpu avx512 cmake cpp cpp17 cpp17-library cuda gpgpu hip hpc hpc-tools nvidia-cuda sandia-national-laboratories scr-2619 simd snl-science-libs vector
Last synced: about 2 months ago
JSON representation
Portably Performant Physical Algebra
- Host: GitHub
- URL: https://github.com/sandialabs/p3a
- Owner: sandialabs
- License: mit
- Created: 2021-04-26T17:02:49.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2023-01-25T16:15:16.000Z (almost 2 years ago)
- Last Synced: 2024-06-14T00:22:26.954Z (7 months ago)
- Topics: amd-gpu, avx512, cmake, cpp, cpp17, cpp17-library, cuda, gpgpu, hip, hpc, hpc-tools, nvidia-cuda, sandia-national-laboratories, scr-2619, simd, snl-science-libs, vector
- Language: C++
- Homepage:
- Size: 645 KB
- Stars: 12
- Watchers: 12
- Forks: 4
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
Portably Performant Physical Algebra
====================================
This is a C++17 library that is meant to support High Performance Computing physics applications.
P3A is meant primarily to help structured grid applications perform well on Intel CPUs, NVIDIA GPUs, and AMD GPUs, but can be used for other applications and can be extended for other hardware.
P3A provides several "layers" of helpful tools:
2. A `p3a::dynamic_array` class that is very similar to `std::vector`, except for an `ExecutionPolicy` template argument that helps it work on GPUs.
3. A few algorithms modeled after C++17's parallel algorithms library such as `p3a::for_each`, `p3a::transform_reduce`, and `p3a::copy`. These include non-standard overloads for iterating over 3D grids.
4. A set of tensor algebra types such as `p3a::symmetric3x3`, `p3a::vector3`, and `p3a::matrix3x3`, with relevant mathematical operator overloads and load/store helpers.
5. A `p3a::quantity` scalar type that adds compile-time units checking and conversion
6. A set of SIMD scalar types `p3a::simd` that directly call CPU vector intrinsics. Currently this is limited to one for Intel AVX-512 and one that is just a scalar.
The ideal design for an application using P3A is to implement a physics code as a series of loops which perform outer loop vectorization, while keeping the C++ source code readable by using tensor types with overloaded mathematical operators.
P3A depends on [Kokkos](https://github.com/kokkos/kokkos) for its parallel algorithms,
and [MPICPP](https://github.com/sandialabs/mpicpp) for its MPI interface.
At Sandia, P3A is SCR 2619.0