ecrc / kblas-gpu
Subset of BLAS routines optimized for NVIDIA GPUs
☆68Updated 2 years ago
Alternatives and similar repositories for kblas-gpu:
Users that are interested in kblas-gpu are comparing it to the libraries listed below
- Next generation library for iterative sparse solvers for ROCm platform☆79Updated this week
- MagmaDNN: a simple deep learning framework in c++☆49Updated 4 years ago
- Next generation LAPACK implementation for ROCm platform☆100Updated this week
- Examples for using SYCL on CUDA☆62Updated last month
- High-performance, GPU-aware communication library☆85Updated 3 months ago
- 🎃 GPU load-balancing library for regular and irregular computations.☆62Updated 10 months ago
- Archived implementation of BLAS using the SYCL open standard. See oneMath for a replacement.☆261Updated 3 months ago
- NUMA-aware multi-CPU multi-GPU data transfer benchmarks☆23Updated last year
- sparse matrix pre-processing library☆81Updated 11 months ago
- Next generation SPARSE implementation for ROCm platform☆120Updated this week
- ☆43Updated 4 years ago
- Intel Data Parallel C++ (and SYCL 2020) Tutorial.☆93Updated 3 years ago
- Reusable software components for ROCm developers☆83Updated last week
- HPCG benchmark based on ROCm platform☆37Updated last month
- ☆28Updated this week
- ☆91Updated 8 years ago
- ☆29Updated 5 years ago
- ROCm SPARSE marshalling library☆67Updated this week
- ☆60Updated 4 months ago
- ROCm Parallel Primitives☆171Updated this week
- Kernel Tuning Toolkit☆59Updated last month
- A simple profiler to count Nvidia PTX assembly instructions of OpenCL/SYCL/CUDA kernels for roofline model analysis.☆50Updated last month
- Generate simple index ranges in C++ and CUDA C++☆39Updated last year
- ROCm Thrust - run Thrust dependent software on AMD GPUs☆107Updated this week
- A GPU benchmark suite for assessing on-chip GPU memory bandwidth☆104Updated 7 years ago
- SYCL Benchmark Suite☆64Updated 2 months ago
- portDNN is a library implementing neural network algorithms written using SYCL☆113Updated 11 months ago
- GTensor is a multi-dimensional array C++14 header-only library for hybrid GPU development.☆36Updated 2 weeks ago
- Efficient SpGEMM on GPU using CUDA and CSR☆52Updated last year
- ☆14Updated 4 years ago