Introduction

What is hypredrive?

hypredrive is a high-level interface for hypre, a package for solving sparse linear systems of equations. It is designed to make configuring and running hypre solvers simple via YAML - supplied either as a file on disk or as an in-memory string from application code. Whether you are a researcher or an application developer, hypredrive offers a lightweight, low-overhead way to access the solvers provided by hypre.

Key Features

  • Encapsulation hides the complexity of the Hypre library by offering an intuitive interface driven by YAML configurations. This allows for straightforward and error-resistant setup, enabling easy adjustments and sharing of solver settings.

  • Prototyping with a variety of solver options. Users can effortlessly compare solver performance and adjust parameters through the YAML configuration, fostering experimentation and optimal solver strategy identification.

  • Testing through an integrated framework to evaluate solvers against a diverse set of predefined linear system problems. Thus ensuring that future hypre developments do not negatively impact solver convergence and performance for these problems.

Getting Started

To get started with hypredrive, you should first ensure that the software is properly installed on your system. For detailed installation instructions, please refer to the Installation section.

Once the installation is complete, familiarize yourself with the YAML configuration reference in the Input Structure (YAML) section. This will provide you with a good understanding of how to configure and run hypredrive-cli for your specific needs.

Here’s an example command to run hypredrive-cli on a single process with a basic configuration file:

$ mpirun -np 1 ./hypredrive-cli input.yml

In this command, input.yml should be replaced with the path to your actual configuration file. You can find input file examples and detailed explanations in the Driver Examples (hypredrive-cli) section.

Contributing

We welcome contributions from the community and are pleased that you’re interested in helping improve hypredrive! This document provides guidelines and information on how you can contribute.

Ways to Contribute

There are many ways to contribute to hypredrive:

  • Reporting Bugs: If you encounter issues or bugs, please report them by opening an issue on our GitHub issues page. Please provide as much detail as possible to help us understand and address the issue.

  • Feature Requests: Are you a developer with ideas for new features or improvements? Feel free to submit them as issues, labeling them as feature requests.

  • Submitting Patches: If you’ve fixed a bug or implemented a feature, you can submit a pull request. Make sure your code adheres to the project’s coding standards and include tests if possible.

If you plan to submit a pull request, before you start, it’s a good idea to get familiar with the following:

  • Project Structure: Understand how the project is organized.

  • Coding Standards: Follow the coding style and guidelines of the project to ensure consistency.

  • Testing: Write and run tests to make sure your changes don’t introduce new issues.

Submitting a Pull Request

  1. Fork the Repository: Start by forking the repository on GitHub.

  2. Clone Your Fork: Clone your fork locally and create a new branch for your contribution.

  3. Make Your Changes: Implement your fix or feature.

  4. Test Your Changes: Ensure your changes pass all tests and don’t introduce new issues.

  5. Commit Your Changes: Commit your changes with a clear, descriptive commit message.

  6. Push Your Changes: Push your changes to your fork on GitHub.

  7. Submit a Pull Request: Open a pull request from your fork to the main hypredrive repository.

Code Review Process

After you submit a pull request, the project maintainers will review your changes. During the review process:

  • Be open to feedback and willing to make revisions.

  • Discuss any suggestions or issues that reviewers bring up.

  • Once your pull request is approved, a maintainer will merge it into the project.