GSoC 25: Parallelization of existing tools

GRASS GRASS developer
1

Hello Grass Team,

I am Mostafa Magdy from Egypt, a fourth-year Computer Engineering student at Cairo University with real-world software development experience through internships at startups like Obelion. I am proficient in C/C++, Python, and JavaScript, and I have completed many projects using these languages. Last year, I made open-source contributions to The Plone Foundation.

I am particularly interested in the OSGEO organization and its projects, especially GRASS, specifically regarding the project for the parallelization of existing tools. I have experience with OpenMP from my courses and have worked with it. I have set up and compiled the project locally; however, I am unsure of the next steps.

I noticed that the issue referenced on the idea page is already closed and merged during GSoC 2024: Issue #2644. Should I choose other tools rather than the ones implemented during GSoC 2024?

Could anyone advise on the project focus for this year?

2

Hi Mostafa,

Welcome! I don’t have a suitable bug ticket tight now, but for the test of skills I was thinking you could try to work on parallelization of v.surf.idw and see how far you can get. Maybe you don’t have to deliver a complete solution, but something we can evaluate you on. Does that sound good?

In terms of the application, there are several tools that would benefit from parallelization. My priority is r.mapcalc and r.proj, they are heavily used and r.proj can be pretty slow. There has been attempts to parallelize them in the past, so these might not be the easiest to start with, but we could come up with other other ones.

Best,
Anna

3

Thanks, @annakrat! I will start working on parallelizing v.surf.idw in a remote branch, and once I make some progress, I’ll mention you so we can discuss it.