Analysing a simulation
To run HBT-HERONS, you always need to provide a path to a parameter file (<PATH_TO_PARAMETER_FILE>). Several examples of
different parameter files are provided in examples/configs. A description of all accepted parameters, their default values, and descriptions can be found in the runtime parameters page.
Warning
Given its history-based approach, HBT-HERONS requires (many) more than one snapshot to produce sensible results. The spacing between outputs should be sufficiently small to capture the formation of subhaloes in isolation, which allows the code to tag their particles for subsequent identification if they become satellites. If the time spacing is too large, any subhalo that forms and becomes a satellite between two consecutive snapshot outputs will not be extracted from the background of the central subhalo that hosts it.
In the limit of analysing a simulation with a single output, HBT-HERONS becomes a glorified centre finder for Friends-of-Friends groups, as no satellites exist.
Analyse all outputs
If you want to analyse all of the outputs of the simulation, the following
command will make HBT-HERONS run from the first output (MinSnapshotIndex) to the
last one (MaxSnapshotIndex) of the simulation:
You can specify the number of OMP threads to use per MPI rank using export OMP_NUM_THREADS=<NUMBER_OMP_THREADS> before
running the command above.
Analyse a subset of outputs
If you are interested in running HBT-HERONS for a subset of the outputs, you can specify the range as follows:
where <START_OUTPUT> is greater or equal than MinSnapshotIndex and <END_OUTPUT> is
less or equal to MaxSnapshotIndex. This command is also used to continue a run that already
started. In such a case, <START_OUTPUT> is the output number
following the last one that was analysed, and <END_OUTPUT> equals MaxSnapshotIndex.
Warning
Adding a value to <END_OUTPUT> is important when restarting, since otherwise HBT-HERONS will set <END_OUTPUT> = <START_OUTPUT>, and hence it will only analyse a single output.
Warning
If you re-run HBT-HERONS on a snapshot that is at higher redshift than your latest available HBT-HERONS catalogue, you will likely need to discard all catalogues that were generated from subsequent snapshots. Please verify that you specify the correct snapshots to analyse when continuing an existing HBT-HERONS run.
For example, if you have catalogues for snapshots 0 1 2 3 4 5 but restart from snapshot 2, you might need to remove catalogues 3 4 5. This happens because the TrackId assigned to each subhalo when it is first resolved (SnapshotOfBirth) varies from run to run, meaning that subhaloes with SnapshotOfBirth = 2 will have a different TrackId value than from the original run. Thus, the TrackId values in catalogues 3 4 5 for those subhaloes will differ from catalogues 2, making the merger trees based on TrackId invalid.