Community projects

[fruzsinaagocs]

Please comment here with the name of the software (linked to the website) you wish to contribute to, and a sentence or two about what it does.

Please link your proposal.md below. (You are encouraged to comment here as soon as you've chosen a package; you can keep working on the proposal.md file.)

---

Example

2️⃣ PETSc: Portable Extensible Toolkit for Scientific computing

PETSc provides parallel algebraic solvers, time integrators, and optimizers for large-scale systems, such as those governed by PDE, with optional tools for managing meshes and discretizing such equations. It offers run-time configuration and rich diagnostics.
proposal.md

---

The above was produced by the following markdown (included in case you're new to markdown)

Details

## :two: [PETSc](https://petsc.org): Portable Extensible Toolkit for Scientific computing
PETSc provides parallel algebraic solvers, time integrators, and optimizers for large-scale systems, such as those governed by PDE, with optional tools for managing meshes and discretizing such equations. It offers run-time configuration and rich diagnostics.
[proposal.md](https://github.com/cu-numpde/fall23-community-jedbrown/blob/main/proposal.md)

[midsorbet]

Trixi.jl

Trixi.jl is a numerical simulation framework written in Julia for conservation laws such as hyperbolic partial differential equations. It offers supports for spectral methods such as Galerkin collocation spectral element method (DGSEM), adaptive mesh refinement (AMR) with support for curved meshes and unstructured quadrilateral meshes.

My proposal.md

Community Software Analysis Proposal

Software: Trixi.jl

Trixi.jl is a numerical simulation framework written in Julia for conservation
laws such as hyperbolic partial differential equations. It offers supports for spectral
methods such as Galerkin collocation spectral element method (DGSEM), adaptive mesh
refinement (AMR) with support for curved meshes and unstructured quadrilateral meshes.

The primary audience for this project are researchers and students. The project is open
to contributions from newer members and the GitHub issues are extensively labelled with
labels such as good first issue, possible student project, and help wanted which
would make it a good fit for me to navigate and understand the code and contribute to the
project.

Stats

Description	Your answer
Repository URL	https://github.com/trixi-framework/Trixi.jl
Main/documentation website	https://trixi-framework.github.io/Trixi.jl/stable/
Year project was started	2020
Number of contributors in the past year	27
Number of contributors in the lifetime of the project	61
Number of distinct affiliations	>10
Where do development discussions take place?	GitHub Issues
Typical number of emails/comments per week?	1-3 issues/week and 1-5 comments/week
Typical number of commits per week?	5-10 (11 commits every week for the last 2 weeks)
Typical commit size	>10 files changed and >100 LoC modified (new feature), ~1 file changed and <100 LoC modified (fixes/improvements)
How does the project accept contributions?	GitHub Pull Requests
Does the project have an automated test suite?	yes
Does the project use continuous integration?	yes
Are any legal/licensing steps required to contribute?	no - MIT License

Install and run

Check the following boxes when complete or add a note below if you
encountered a problem.

• I have installed the software

• I have run at least one example
  

• I have run the test suite

• The test suite passes

  Test Summary:  | Pass  Total     Time
  Trixi.jl tests |    1      1  5m37.3s
  337.393766 seconds (458.30 k allocations: 25.888 MiB, 0.10% compilation time)
       Testing Trixi tests passed
  

Notes/concerns/risks

Contributing a new feature to the library itself would be really difficult with my current level of PDE knowledge.

Note on copyright

Students retain copyright on any work done in completion of a CU
course, so you are authorized to sign a contributor license
agreement (CLA),
affirm a developer's certificate of
origin (DCO),
etc. If you have concerns about this, please note them and/or reach
out to Jed directly.

BTW, I couldn't find the `Settings` option in my repository

[fruzsinaagocs]

@midsorbet Thanks for flagging the visibility issue. It turns once an assignment is created, your privileges (such as changing your repo's visibility) cannot be changed, so let's just go with private repos for now.

[mohadesekhoshnevisan99]

🔢 HONEE: High-Order Numerical Electromagnetics Environment

HONEE is a high-performance simulation toolkit for solving Maxwell’s equations using spectral-element methods.
It supports complex geometries and high-order accuracy, ideal for wave propagation problems.
The software is written in modern C++ and optimized for parallel execution.

proposal.md

[grahammauer]

NumPy is a fundamental scientific computing package in Python. It includes multidimensional array objects, fast operations and broadcasting on arrays, as well as linear algebra functionalities. Most of its core processes are written in C and it uses BLAS and LAPACK for its linear algebra functionalities.

proposal.md

[tangscott959]

SciPy is a Python library for scientific computing that builds on NumPy and provides tools for optimization, integration, interpolation, linear algebra, statistics, and more. It is widely used in data science, engineering, and scientific research for performing complex mathematical computations efficiently.

proposal.md

[Zf98ai]

libMesh provides a finite element framework for the numerical simulation of partial differential equations using unstructured meshes in 1D, 2D, and 3D. It supports both steady and transient simulations, and enables scalable parallel computations with adaptive mesh refinement. Designed for flexibility and extensibility, it allows researchers to focus on physical modeling while leveraging robust solvers through interfaces with PETSc, Trilinos, and optional eigenvalue solvers like SLEPc.

proposal.md

[lkwargs]

NeuralPDE.jl is a solver package which consists of neural network solvers for PDEs based on physics-informed neural networks (PINNs).
proposal.md

[fruzsinaagocs]

Thanks everyone, we'll present in this order tomorrow:

1. Lawrence
2. Mohadeseh
3. Graham
4. Scott
5. Fengyu
6. Lekai