Run your first MESA HMS-HMS binary simulation with POSYDON

If you haven’t done it already, export all the relevant POSYDON and MESA environment variables.

Tip: create a file called .bash_POSYDON_MESA with the following content (please adapt the PATHs to your system):

export PATH_TO_POSYDON=~/software/POSYDON/
export PATH_TO_POSYDON_DATA=~/software/POSYDON/data/

export MESA_DIR=~/software/mesa-r11701_witheoschange_andwithreverseMTchange_fiximplicitmdot
export OMP_NUM_THREADS=4
export MESASDK_ROOT=~/software/mesasdk
source $MESASDK_ROOT/bin/mesasdk_init.sh

so that every time you open a new terminal you can just run:

source .bash_POSYDON_MESA

If you haven’t done it already, please download the development branch of the POSYDON-MESA-INLISTS submodule, see Architecting MESA Simulation Grids with POSYDON. This is the branch that contains the simulation properties of all POSYDON MESA grids.

Creating the Initialization File for the POSYDON MESA Submission Script

You can now copy in your working directory the API ini file configuration of HMS-HMS binaries, which is stored on Github as part of the POSYDON repository. The cell below will download the file for you.

Note: in the POSYDON-MESA-INLISTS submodule we have more example ini files at POSYDON-MESA_INLISTS/r11701/running_scripts/. For each grid we support two ini file configured to the Northwestern and UNIGE HPC clusters.

[ ]:

import urllib.request

# Download the grid_params.ini file from GitHub
url = "https://raw.githubusercontent.com/POSYDON-code/POSYDON/refs/heads/main/grid_params/grid_params.ini"
urllib.request.urlretrieve(url, './HMS-HMS.ini')
print("Download completerd")

Open the file and edit the following parameters:

  • user=your_username (can be skipped if not run via SLURM)

  • partition=the_slurm_partition_to_run_on (can be skipped if not run via SLURM)

  • account=your_slurm_account (can be skipped if not run via SLURM)

  • email=your_email (can be skipped if not run via SLURM)

  • posydon_github_root=your_path_to_POSYDON

  • grid=your_grid_csv_file (will get created in the next section)

The scenario parameter of the ini file is the main parameter that controls the simulation setup. This parameter let’s us decide which simulation we want to run. The syntax is the following:

  • the first argument is the name of the submodule posydon which points to the POSYDON-MESA-INLISTS submodule (other options are available, e.g. user which points to a private submodule)

  • the second argument is the name of the branch and commit of the submodule we want to use connected by a dash -. In this case we want to use the development branch of the submodule and it’s latest commit, so we write development-c1f75acacc3bd645197c5d3e920a121e94d95a80

  • the third argument is the name of the simulation grid we want to run, in this case we want to run a HMS-HMS grid. The name of the grid is HMS-HMS.

To summarize, you should have scenario = ['posydon', 'development-c1f75acacc3bd645197c5d3e920a121e94d95a80', 'HMS-HMS']

The zams_file points to the ZAMS model used to generate the HMS stars in the binary system. POSYDON v2.0.0 supports 8 different metallicities, here we would like to use the 0.1Zsun POSYDON MESA ZAMS model ${posydon_github_root}/grid_params/POSYDON-MESA-INLISTS/r11701/ZAMS_models/zams_z1.42m3_y0.2511.data

The last parameter of this file is grid and points to the csv file containing the initial conditions of the grid. In this case we want to run a HMS-HMS grid consisting of one system, so we set grid = grid_test.csv and generate such file specifying the metallicity, the two star masses and the initial orbital period.

Creating the Initial Simulation Points CSV File

Note: a notebook showing you how all the grid points of v2.0.0 are generated is available in the POSYDON-MESA-INLISTS submodule at $PATH_TO_POSYDON/grid_params/POSYDON-MESA-INLISTS/r11701/running_scripts/parameter_space_v2/create_csv.ipynb

[ ]:

import csv

with open('./grid_test.csv', 'w', newline='') as file:
    writer = csv.writer(file)
    writer.writerow(['initial_z','Zbase','m1','m2','initial_period_in_days'])
    writer.writerow([0.00142, 0.00142, 30., 21., 10.])

The above script will create a .csv file for the parameters of a MESA ‘grid’ with one row, i.e. practically with one binary system. You should now have the following files in your working directory:

[ ]:

!ls

Let’s now use POSYDON to set up the MESA simulation. In your terminal run

posydon-setup-grid --grid-type fixed --inifile HMS-HMS.ini --submission-type slurm

The following files will be created:

[ ]:

!ls

Running the MESA Model and Exploring the Simulation Output

You are now ready to submit the simulation to the cluster with the following command:

[ ]:

!./run_grid.sh

[ ]:

!squeue -u kruckow

After the job is finished, we can check the output files:

[ ]:

!ls

[ ]:

!ls Zbase_0.0014_m1_30.0000_m2_21.0000_initial_z_1.4200e-03_initial_period_in_days_1.0000e+01_grid_index_0

[ ]:

!gzip -cd Zbase_0.0014_m1_30.0000_m2_21.0000_initial_z_1.4200e-03_initial_period_in_days_1.0000e+01_grid_index_0/out.txt.gz | head -n 320

The above output is standard output of the beginning of a MESA binary run. Congratulations, you now know how to leverage POSYDON to run a grid of MESA simulations on a computer cluster!