-
Notifications
You must be signed in to change notification settings - Fork 217
Parameter Files
You are here: Home > PIConGPU User Documentation > Parameter Files
Parameter files, *.param
placed in include/simulation_defines/param/
are used to set all compile-time options for a PIConGPU simulation. This includes most fundamental options such as numerical solvers, floating precision, memory usage due to attributes and super-cell based algorithms, density profiles, initial conditions etc.
High-performance hardware comes with a lot of restrictions on how to use it, mainly memory, control flow and register limits. In order to create an efficient simulation, PIConGPU compiles to exactly the numerical solvers (kernels), physical attributes (fields, species) for the setup you need to run which will furthermore be specialized for a specific hardware.
This comes at a small cost: when one of those settings is changed, you need to recompile. Nevertheless, wasting about 5 minutes compiling on a single node is nothing compared to the time you save at scale!
All options that are less or non-critical for runtime performance, such as specific ranges observables in Plugins or how many nodes shall be used can be set in run time configuration files (*.cfg) and do not need a recompile when changed.
If you use our configure
script wrappers, you do not need to set all available parameter files since we will add the missing ones with sane defaults. Those defaults are:
- a standard, single-precision, well normalized PIC cycle suitable for relativistic plasmas
- no external forces (no laser, no initial density profile, no background fields, etc.)
See our new online documentation for a per param file description.
When editing specific parameter files, you can follow along the arrows below which represent roughly the dependencies (or hierarchy) of the individual files.
So easy, isn't it? ;)
All wiki entries describe the dev branch. Features may be different in the current master branch.
Before you start please read our README!
PIConGPU is a scientific project. If you present and/or publish scientific results that used PIConGPU, you should set a reference to show your support. Our according up-to-date publication at the time of your publication should be inquired from:
The documentation in this wiki is still not complete and we need your help keeping it up to date. Feel free to help improving this wiki!