The predefined configuration files¶
Configuring the collision setups¶
There are several complete handler configuration files prepared for the user to run the hybrid handler in its entirety for different collision systems and energies. They follow the setup chosen in Schäfer et al.: Eur.Phys.J.A 58 (2022) 11, 230. However, the finer hydrodynamic grid is by default commented out and has to be uncommented to reproduce the same setup as in the publication. The shear viscosities applied are taken from Karpenko et al.: Phys.Rev.C 91 (2015) and the longitudinal and transversal smearing parameters are adjusted to improve agreement with experimental data. The supported collision systems are listed in the following table.
System | \(\mathbf{\sqrt{s_{NN}} \;\: \{GeV\}}\) | System | \(\mathbf{\sqrt{s_{NN}} \;\: \{GeV\}}\) |
---|---|---|---|
Au + Au | 4.3 | Pb + Pb | 6.4 |
Au + Au | 7.7 | Pb + Pb | 8.8 |
Au + Au | 27.0 | Pb + Pb | 17.3 |
Au + Au | 39.0 | Pb + Pb | 2760.0 |
Au + Au | 62.4 | Pb + Pb | 5020.0 |
Au + Au | 130.0 | ||
Au + Au | 200.0 |
They can be found in configs/predef_configs folder and the user is only required to insert the paths of the executables in the individual software sections.
They can be executed in the standard manner using -c
option in the execution mode of the handler.
./Hybrid-handler do -c configs/predef_configs/config_AuAu_4.3.yaml
Be aware of computational cost!
The default predefined setup is meant to be executed on a computing cluster, we do not recommend executing it locally. If you want to test a simple setup on your local machine, refer to the test setup explained here below.
Tradeoff between runtime and precision
Although SMASH-vHLLE-hybrid conserves in average all charges and almost all energy, this is strongly dependent on the grid in \(\eta\) direction. Therefore, there is a tradeoff between the runtime and memory consumption of the hydrodynamic stage and conservation. The default in the configuration tries to strike a balance, as there is less than 10% loss of energy for central collisions from \(\small\sqrt{s} = 4.3\;\mathrm{GeV}\) to \(\small\sqrt{s} = 200\;\mathrm{GeV}\). The original publication Schäfer et al.: Eur.Phys.J.A 58 (2022) 11, 230 used a considerably finer grid, resulting in around 8 times higher runtime and memory consumption, but also less than 7% loss for small energies and almost perfect conservation for high energies.
Running a test setup¶
To test the functionality and to also run it on a local computer, there is the possibility to execute a test setup of the hybrid handler, which is a Au+Au collision at \(\small\sqrt{s} = 7.7\;\mathrm{GeV}\). The statistics are significantly reduced and the grid for the hydrodynamic evolution is characterized by large cells, too large for a realistic simulation scenario.
Don't use this setup for production!
This test setup is only meant to be used in order to test the functionality of the framework and the embedded scripts, but not to study any realistic physical system.
To execute the test, a predefined configuration file is prepared in the same predef_configs folder. Again, the user needs to insert the paths to the executables in all the software sections.