A highly scalable, MPI-based 3D stencil computation engine with isovalue crossing detection, optimized for HPC clusters.
From-scratch Canny edge detector in C++/Qt, parallelized with MPI — demonstrates why naive domain decomposition corrupts stencil computations at strip boundaries, and how halo/ghost regions fix it.
2D Heat Diffusion HPC Simulator. Performance and speedup comparison across Sequential C++, OpenMP, MPI, and CUDA (GPU) implementations.
Defensive publication: executable reference for conflict-free streaming stencils with single-port banked SRAM, multicast delivery, and ping-pong buffering.
To associate your repository with the stencil-computation topic, visit your repo's landing page and select "manage topics."