Flood2D HPC Simulation using MPI

Overview

Parallel C++ simulation of 2D flood propagation using MPI.

Flood2D is a research-oriented scientific computing project implementing a two-dimensional flood simulation using MPI-based parallelization in C++.

Key Features

• 2D grid-based numerical simulation
• Domain decomposition across MPI processes
• Parallel execution using MPI (message passing)
• Deterministic and reproducible simulation results
• Designed for execution on HPC clusters and multi-core systems

Technical Approach

• C++ implementation with explicit memory management
• MPI for inter-process communication and synchronization
• Structured grid decomposition across ranks
• Boundary data exchange between neighboring subdomains
• Scalable execution model suitable for distributed environments

Technologies

• Language: C++
• Parallel Computing: MPI
• Execution Environment: Linux, HPC clusters
• Build & Run: Standard MPI toolchain (mpic++, mpirun)

Project Structure

src/ -> source code
results/ -> outputs
assets/ -> figures / performance plots

Example Execution

Compile: mpic++ flood2d.cpp -o flood2d

Run locally: mpirun -np 4 ./flood2d

Example SLURM job script: #!/bin/bash #SBATCH --job-name=flood_sim #SBATCH --nodes=2 #SBATCH --ntasks-per-node=16 #SBATCH --time=02:00:00

module load mpi mpirun ./flood2d

Research Software Engineering Context

This project was developed to demonstrate parallel scientific software design, HPC execution models, and reproducible research code.

It serves as a reference example for Research Software Engineer and scientific computing roles where scalable computation and collaboration with domain scientists are essential.

Skills Demonstrated

• C++ scientific programming
• MPI-based parallel computing
• Domain decomposition
• HPC execution workflows
• Research software engineering

Authorship & Contributions

This project was developed in a collaborative research environment. The overall simulation framework and architectural design were created as part of a broader research effort, while I contributed to the implementation, MPI-based parallelization, integration, execution workflow, and research-oriented adaptation of the codebase.

This repository is shared to demonstrate experience with scientific computing, MPI-based parallel programming, and research software engineering practices.

Dr. Rajesh Manicavasagam
Research Software Engineer | Scientific Computing | HPC (MPI, CUDA, SLURM)

GitHub: https://github.com/rmanicav
Google Scholar: https://scholar.google.com/citations?user=2XswkUcAAAAJ