Parallel Stencil Computation
High-performance, hybrid MPI+OpenMP implementation of a 2D stencil heat diffusion solver, explicitly optimized for the Leonardo DCGP cluster (Intel Sapphire Rapids).
Parallel ComputingPerformance OptimizationHPC
CMPIOpenMPSLURM
This project implements a numerical simulation of heat diffusion on a 2D plate using a stencil computation method, optimized for the Leonardo DCGP cluster (CINECA).
For a deep dive into the implementation details, compilation flags, and performance analysis, please refer to the technical documentation located in the GitHub repository (`docs/` folder). There you can find the following sections:
- Parallelization Strategy: Detailed explanation of the domain decomposition, MPI communication, and OpenMP integration.
- Compilation Guide: In-depth look at the Makefiles, compiler flags, and optimization levels.
- Leonardo Cluster Specs: Technical specifications of the CINECA Leonardo DCGP partition used for testing.
- Scalability Analysis: Strong and weak scaling results, including performance graphs and overlap efficiency analysis.
Collaborators
Solo project