Getting Started

ENGR 570 - Software Development for Research

Kyle Niemeyer

Oregon State University

2025-01-06

Overview of class and introductions

What this class is for:

Learning computational skills to be successful in modern research
Developing a software package from scratch for your research

What this class isn’t for:

Learning how to program in Python
Those who aren’t comfortable doing open-ended, self-directed work.

Motivation

Nearly all research relies on software—even experimental—but researchers don’t get trained in best practices in the same way as experimental methods.

What are we going to learn about?

Computational environment management
Version control
Collaborative software development
Structuring programs
Working with files
Improving performance, parallelization
Reproducible research
Writing about software, documentation
Sharing software openly, copyright

Anything missing?

Introductions

Who are you? Quick introductions

Majors / research areas?
Experience level in Python / programming
Experience level with command line
Mac, Linux, or Windows?

Who am I?

Photo of Kyle Niemeyer

Dr. Kyle Niemeyer, Associate Professor of Mechanical Engineering

Background: computational combustion modeling, fluid dynamics

Languages: Matlab -> Fortran -> C/C++/CUDA -> Python

Course Resources

Online resources

Syllabus: on Canvas
GitHub organization: https://github.com/SoftwareDevEngResearch
Other resources: linked on Canvas
Slack chat room (linked on Canvas)

Textbook

Effective Computation in Physics book

http://physics.codes/

Chapters 1–6 should be review

Project

Most of your work in the class will be centered around a project developing a new software package for your research.

Assignments will focus on development stages for your software package.

First assignment: project proposal, due January 17.