Binning Plugin TODOs and Roadmap

[ikbuibui]

TODOs for the flexible Binning plugin

• [ ] Add flexible accumulation methods - Currently only addition is supported
• [ ] Auto binning ranges and splitting - Automatically and dynamically in the simulation choose binning range min, max and number of bins
• [ ] Particle Filtering - Bin/Don't bin particles matching certain criteria
• [ ] Add more axes
  1. Log Axes
  2. Generic Axes defined with edges
  3. LogLinLog Axes
  4. String Axes
• [ ] Reduce (MPI) the binning histogram buffer in place
• [ ] Profile the binning
• [ ] Bin a particle multiple times - Useful for RNG based functors, say to generate distributions
• [ ] Option to only bin particles which are leaving (https://github.com/ComputationalRadiationPhysics/picongpu/issues/4946#issuecomment-2180457170)
• [ ] Add ability to apply post processing functor on the histogram before doing a data dump
• [ ] Discuss sane defaults, default examples and pre-define functors for common quantities
• [ ] Bin fields (#4972)
• [x] Restart and Checkpoint Support (#4766)
• [x] Add ability to disable overflow bins (#4770)

[psychocoderHPC]

@ikbuibui If I do not miss something then the binning plugin is missing the OpenMPD annotation if a value is including the weighting and how it scales with the weighting. The only OpenPMD anotation the user can pass the openPMD unit description.

[ikbuibui]

As discussed in the office, the weighting is a particle property and not a property of bins, so the the weighting annotation for bins makes no sense. If the user needs special treatment of weighted quantities it can be done inside the functors.