A high-performance implementation of Sparse Matrix-Vector Multiplication in C++ with serial, parallel (OpenMP), and GPU-accelerated (CUDA) versions, demonstrating the performance benefits of parallelism across different architectures.
serialgpuoptimizationopenmpparallel-computingcudaefficiencymethodologycsrspmvsparse-matrixspeedupworkloadload-balancingfostercompressed-sparse-rowproblem-size
-
Updated
Dec 29, 2025 - C++