AMDResearch / DAGEE
Directed Acyclic Graph Execution Engine (DAGEE) is a C++ library that enables programmers to express computation and data movement, as task graphs that are scheduled concurrently and asynchronously on both CPUs and GPUs.
☆45Updated 3 years ago
Alternatives and similar repositories for DAGEE:
Users that are interested in DAGEE are comparing it to the libraries listed below
- SYCL Reference Manual☆27Updated 10 months ago
- A simple profiler to count Nvidia PTX assembly instructions of OpenCL/SYCL/CUDA kernels for roofline model analysis.☆50Updated this week
- Asynchronous Task and Memory Interface, or ATMI, is a runtime framework and programming model for heterogeneous CPU-GPU systems. It provi…☆66Updated last year
- ☆44Updated this week
- mallocMC: Memory Allocator for Many Core Architectures☆55Updated last month
- AMD’s C++ library for accelerating tensor primitives☆38Updated last week
- Copy-hiding array abstraction to automatically migrate data between memory spaces☆106Updated last week
- SYCL Benchmark Suite☆64Updated last month
- Compiler agnostic metaprogramming library providing concepts, type operations and tuples for C++ and cuda☆84Updated last week
- Next generation LAPACK implementation for ROCm platform☆99Updated this week
- SYCL Conformance Tests☆68Updated last week
- ROCm Thrust - run Thrust dependent software on AMD GPUs☆106Updated this week
- Generate simple index ranges in C++ and CUDA C++☆39Updated last year
- Distributed ranges is a generalization of C++ ranges for distributed data structures.☆49Updated last week
- ☆56Updated last week
- Bandwidth test for ROCm☆54Updated last week
- Reusable software components for ROCm developers☆83Updated this week
- High-performance, GPU-aware communication library☆84Updated 2 months ago
- A unified framework across multiple programming platforms☆36Updated 9 months ago
- DLA-Future☆70Updated last week
- ☆11Updated last month
- rocSHMEM intra-kernel networking runtime for AMD dGPUs on the ROCm platform.☆62Updated this week
- RAJA Performance Suite☆118Updated last week
- GPUDirect Async support for IB Verbs☆104Updated 2 years ago
- ROCm SPARSE marshalling library☆67Updated last week
- pika is a C++ tasking library built on std::execution with fibers, CUDA, HIP, and MPI support.☆71Updated this week
- Autonomic Performance Environment for eXascale (APEX)☆44Updated this week
- Matrix multiplication on GPUs for matrices stored on a CPU. Similar to cublasXt, but ported to both NVIDIA and AMD GPUs.☆30Updated 3 months ago
- ROCm Systems Profiler☆16Updated this week
- Unit benchmarks of CUDA event APIs.☆17Updated 10 months ago