ASTR 427 METHODS OF COMPUTATIONAL ASTROPHYSICS
This is a hands-on course to learn methods for numerically
solving problems that arise in astrophysics. Some programming
experience is required. Teaching programming is not the primary goal,
nevertheless it is hoped that techniques for writing and debugging
programs will be a useful side effect.
Numerical Recipes by Press et al. (3rd edition is
best, but earlier editions are OK)
For tips on good practices for writing scientific software, I
recommend Writing Scientific Software: A Guide to Good Style by
Oliveira and Stewart. This book is somewhat language neutral. For a
book that specifically focusses on object oriented C++ (not what I
recommend for this course), I
suggest Scientific and Engineering C++: An Introduction with
Advanced Techniques and Examples by Barton and Nackman.
- Interpolation and extrapolation
- Ordinary Differential Equations
- Linear Algebra
- Root Finding
- Monte-Carlo methods
- Data structures and algorithms
- Parallel techniques
The grades will be entirely based on 5 programming assignments. Correctness of
results, program performance, and readability will be used as grading
The student will understand performance issues and limitations of
floating point arithmetic on computers.
The student will be able to express and solve several types of
astrophysics calculations within a computer program.
The student will have skills to assess correctness of computed results.
Paper copies are OK, but assignments must be turned in electronically,
preferably via email to the instructor's address below.
Github repositories are also welcome. My github
handle is trquinn.
Hyak GPU node access:
See the Hyak GPU HOWTO
Instructor Contact Information
Office: PAB 380, email: email@example.com, phone: 206-685-9009, and fax: 206-685-0403