Adding a Benchmark

This guide is intended for application developers who want to add a benchmark so that it can be run with Benchpark.

Create a New Benchmark

The following system-independent specification is required for each ${Benchmark1} in benchpark/repo/${Benchmark1}:

• package.py is a Spack specification that defines how to build and install ${Benchmark1}.

• application.py is a Ramble specification that defines the ${Benchmark1} input and parameters.

During benchpark setup the user selects ${Benchmark1} to run as the following:

benchpark setup </output/path/to/experiments_root> </output/path/to/system_root> </output/path/to/workspace>

By default, Benchpark will use ${Benchmark1} specifications (application.py and package.py) provided in the Spack and Ramble repos. It is possible to overwrite the benchmark specifications provided in the Spack and Ramble repos; see Frequently Asked Questions for details.

Validate a Benchmark

Now that the benchmark has been created/updated you need to configure at least one experiment, and then validate it, see Adding an Experiment.