CHAI
Copy-hiding array abstraction to automatically migrate data between memory spaces (by LLNL)
RAJA
RAJA Performance Portability Layer (C++) (by LLNL)
CHAI | RAJA | |
---|---|---|
1 | 2 | |
103 | 440 | |
1.9% | 1.6% | |
7.3 | 9.7 | |
8 days ago | 34 minutes ago | |
C++ | C++ | |
BSD 3-clause "New" or "Revised" License | BSD 3-clause "New" or "Revised" License |
The number of mentions indicates the total number of mentions that we've tracked plus the number of user suggested alternatives.
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
CHAI
Posts with mentions or reviews of CHAI.
We have used some of these posts to build our list of alternatives
and similar projects. The last one was on 2021-03-19.
-
Guidelines for using raw pointers in modern C++ and GPUs
I believe most of these performance portability libraries have associated memory manager projects. For example, RAJA has CHAI. I would use those managed array classes.
RAJA
Posts with mentions or reviews of RAJA.
We have used some of these posts to build our list of alternatives
and similar projects. The last one was on 2022-04-24.
-
Cuda application question
Since the ability to use C++ parallel algorithms on the GPU is a relatively new thing, some applications have used other C++ abstraction libraries instead, such as Kokkos (https://kokkos.org/) and RAJA (https://github.com/LLNL/RAJA). These both have multiple backends that support GPUs and CPUs without needing to change your application code.
- Kokkos C++ Performance Portability Programming EcoSystem
What are some alternatives?
When comparing CHAI and RAJA you can also consider the following projects:
CuTeLib - CUDA Template Library provides simple, typesafe, performant constructs for C++ CUDA projects
kokkos - Kokkos C++ Performance Portability Programming Ecosystem: The Programming Model - Parallel Execution and Memory Abstraction
Umpire - An application-focused API for memory management on NUMA & GPU architectures
mfem - Lightweight, general, scalable C++ library for finite element methods
ArrayFire - ArrayFire: a general purpose GPU library.
cuda-samples - Samples for CUDA Developers which demonstrates features in CUDA Toolkit
Vc - SIMD Vector Classes for C++