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-benchmarks.sh
./run-benchmarks.sh polar
where arguments are subdirectories of the current one
- Clone the repository to your local machine
git clone https://github.com/phantomSPH/phantom-benchmarks
- Make a new directory
cd phantom-benchmarks
mkdir mybench
cd mybench
- Add the phantom input file with ".in.s" as the file extension
cp mybench.in mybench.in.s
- Add the reference dump file to which results should be compared, with ".ref" as the file extension
cp mybench_00001 mybench_00001.ref
-
Ensure your calculation is SHORT (runs in < 2 minutes) but REPRESENTATIVE, e.g. with typical particle numbers and timestep ranges
-
Verify that the test passes by running
./run-benchmarks mybench
-
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/)
-
Create a file named
.hashlist
, which points to the files that need to be automatically downloaded at run time. Generate the SHA1 checksums for your files usingshasum
(orsha1sum
, depending on your system). For example,shasum mybench_00000
. Your file should look like this:
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx mybench_00000
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx mybench_00001.ref
- 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