Phantom benchmark suite

Benchmarking and performance suite for the phantom code http://github.com/danieljprice/phantom

Tests in this repo are run nightly and checked for speed and are used for performance evaluation/code scaling tests. Results are currently published to http://phantomsph.bitbucket.io/nightly/opt/

Run all benchmarks

./run-benchmarks.sh

Run selected benchmarks

./run-benchmarks.sh polar

where arguments are subdirectories of the current one

How to add benchmark problems to this suite

1. Clone the repository to your local machine

git clone https://github.com/phantomSPH/phantom-benchmarks

2. Make a new directory

cd phantom-benchmarks
mkdir mybench
cd mybench

3. Add the phantom input file with ".in.s" as the file extension

cp mybench.in mybench.in.s

4. Add the reference dump file to which results should be compared, with ".ref" as the file extension

cp mybench_00001 mybench_00001.ref

5. Ensure your calculation is SHORT (runs in < 2 minutes) but REPRESENTATIVE, e.g. with typical particle numbers and timestep ranges
6. Verify that the test passes by running

./run-benchmarks mybench

in the root directory 7. Open an Issue to make a request for the dump files to be copied to the web server (http://data.phantom.cloud.edu.au/data/benchmarks/) 8. Paste the checksums into a file named .hashlist in the test directory, for example

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx mybench_00000
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx mybench_00000.ref

These files are automatically downloaded at run time.

6. Commit and push the benchmark to the repository, excluding the dump files (to avoid bloating the repository)

git add mybench.in.s SETUP .hashlist
git commit -m 'added benchmark mybench'
git push