SMaT can be used to accelerate many workloads in scientific computing, large model training, inference, and others. Requirements Hardware We run our experiments on the Swiss National Computing Center’s Ault compute cluster. Each node is equipped with a single NVIDIA A100-SXM4-40GB GPU, and AMD...
GE-SpMMis a fast CSR-based CUDA kernel of sparse-dense matrix multiplication (SpMM), designed to accelerate GNN applications. Get started git clone --recursive https://github.com/hgyhungry/ge-spmm.git Kernel performance Prerequisites CUDA toolkit 10.1 Compilation source compile.sh The script sho...