Magnetizer
This code takes the output of the Galform semi-analytic model of galaxy formation (SAM) and computes radial dependent ISM properties and magnetic field for each simulated galaxy. The magnetic field is obtained by numerically solving the galactic dynamo equation throughout history of each galaxy.
For more details about the physical model, we refer the reader to the original code paper.
Quick install and run guide
Galaxy Magnetizer requires the following libraries to run:
Once they are installed (see Dependencies for building
instructions), the code can be compiled by using make prod or make test, for
a production or test (debugging and backtracing enable) run, respectively.
For building using multiple processors, the command
make -j <number of processors> should work.
The code can be run using mpi:
export HDF5_USE_FILE_LOCKING=FALSE
mpirun ./Magnetizer.exe <parameters_file>The parameter file must include the path to the HDF5 input file, containing the time-evolving galaxy properties. The first line sets an environment variable which disables file locking for HDF5. This is used to allow multiple processes to read the input and output files simultaneously (while the writing is exclusively done by the root/master process).
The input file can be generated using the scripts/prepare_input.py script. Parameters not specified in the parameters file are set to their default values, thus the minimal parameters file is
&global_pars
input_file_name = "sam_input.hdf5"
/An example parameters file can be found in the
example/example_global_parameters.in.
In the same directory, there is an example input file,
example/example_SAM_input.hdf5.
The Magnetizer can also be run in the single galaxy mode (which is particularly useful for debugging) by simply specifying the galaxy number i.e.
./Magnetizer.exe <parameters_file> <igal> [-f]note that a full output file will be generated, but containing only this galaxy.
The option -f allows one to force re-run a particular galaxy.
Magnetizer comes with a range of Python modules and scripts which can be used for
- preparing the HDF5 input files
- diagnostic of finished runs
- visualization and analysis tasks
The python code will depend on the following
Preparing an input files
Input files can be generated from a Galform (and later other SAMs/sims) run using the prepare_input.py. For information about this command's usage/arguments, please check
./python/prepare_input.py --helpAn example Galform output can be found at scripts/test_SAM_output/galaxies.hdf5.
MagnetizerRun objects
To facilitate any analysis tasks, one can use the MagnetizerRun python object to interact with the input and output. These objects will only load information of galaxies which were completed (i.e. galaxies which did not run either because of error or because Magnetizer was prematurely interrupted are automatically ignored).
run = magnetizer.MagnetizerRun(input_path='example/example_input.hdf5',
output_path='example/example_output.hdf5')It is also possible to supply lists of filenames to input_path and output_path (for example in the case where one is working with various subvolumes of a large simulation, each of them stored in a separate file). The MagnetizerRun object will concatenate the data of the different files in the lists whenever this is needed.
MagnetizerRun objects contain several useful attributes, as
zs = run.redshifts
times = run.times
number_of_galaxies = run.ngals
number_of_grid_points = run.ngridUsing this object, for example, one can load the azimuthal component of the field, $B_\phi$, at $z=0$, for all the galaxies using:
Bp_all_galaxies = run.get('Bp', 0.0)where Bp_all_galaxies will be a (number_of_galaxies)x(number_of_grid_points) array.
One can also load all the data (i.e. from all redshifts) for a specific galaxy using
Bp_gal_answer = run.get_galaxy('Bp', 42)in this case Bp_gal_answer will be a (number_of_grid_points)x(number_of_redshifts) array.
Contact
Please use Github's tools or email Luiz Felippe S. Rodrigues if you find any problem.