Scaling-Tester

This Project encompases a collection of scripts that automate the processes of runing Scaling Tests with TRUST/TrioCFD on a given datafile.

This Project currently uses "sbatch" command (slurm) to run each test. It is posible to change this scripts functionality by replacing/changeing some its modules.

This Project contains the following modules (scripts):

• scaling_tester.py
• run_test.py
• gather_results.py
• parse_output.py
• refine_mesh.py

 

scaling_tester:

Runs full scaling tests on the given datafile.

usage:

• -h, --help     Show this help message and exit.
• -df DATA_FILE Path to the data file to be executed.
• -mp MAX_PROCS Maximal number of processes (the tests will be run for each power of 2 which is lower or equal to MAX_PROCS).
• -t TIMES      Number of times each test will be executed.
• -r             Should the mesh be refined.

 

run_test:

Runs the given test with a specific amount of processes "TIMES" times.

usage:

• -h, --help   Show this help message and exit.
• -tn TEST_NAME The name of the test to be executed.
• -td TEST_DIR   The diretory contaning the tests data files.
• -t TIMES      Number of times each test will be executed.
• -np N_PROCS    Number of processes to run the test with.

gather_results:

Gathers the test results form all output files created by runinig the scaling_tester, and dumps them to a CSV file called "test_results.csv".

usage:

• -h, --help Show this help message and exit.
• -d DIR     Path to the directory created by scaling tester containing the test results.

parse_output:

Parses a given output file and return a dictionary containing its results.

usage:

• -h, --help       Show this help message and exit.
• -of OUTPUT_FILE path to the out file to be refined.

 

refine_mesh:

refines a given datafiles mesh by x4 or x8 elements.

usage:

• -h, --help           Show this help message and exit.
• -df DATA_FILE       Path to the data file to be refined.
• -rt REFINEMENT_TYPE Refinement type 4 or 8.