Skip to content

Latest commit

 

History

History
188 lines (141 loc) · 6.38 KB

FccSoftwareFramework.md

File metadata and controls

188 lines (141 loc) · 6.38 KB

FCCSW

Contents:

Overview

The FCC software framework is based on Gaudi, with a simple event data model (EDM) library called PODIO (Plain Old Data Input / Output). The code is managed with Git, and built using CMake.

Git guide for FCC

The code for both the FCC software and Gaudi is managed on Git repositories. We have set up a short Git guide for people who are not familar with Git, and where we explain the recommended way to contribute code to the FCC software.

All new developers should read this guide.

CMake

The build process of the FCC software is managed with CMake. Here is a short CMake guide{.twikiLink} for FCC developers.

Installation

Log in an lxplus6 machine and follow these instructions .

Example process

An example configuration file is provided in simple_workflow.py . This configuration file allows to execute the following tasks for each event:

  • read an HepMC text file
  • convert the HepMC particles to a collection of EDM ParticleCollection (also see the MCParticleCollection doxygen )
  • read the EDM Particles (also see the MCParticle doxygen ) and use fastjet to cluster them into EDM Jets .
  • write an EDM root file

Get a source HepMC text file (1000 Z production events in an e+e- collider at sqrt(s) = 91 GeV ):

cp /afs/cern.ch/user/c/cbern/public/pythia_Z_91_hepmc.dat example_MyPythia.dat

Run Gaudi on this configuration file:

./run gaudirun.py simple_workflow.py

Please have a detailed look at simple_workflow.py .

Exercises

1- Modifying an existing algorithm

  • locate the C++ code of the jet clustering algorithm (maybe do a recursive grep for jet or clustering?)

  • add a printout for the reconstructed jets in the execute function of the jet clustering algorithm.

  • recompile and run again:

       cd $FCCSW
   make

Note how the total energy is generally way too high. It is due to the fact that both stable and unstable particles are clustered into jets, hence a double counting of the energy. In exercise 3, we will see how to filter the generated particles to send only the stables ones to jet clustering.

2- Changing an algorithm parameter

Open JetClustering.cpp , and have a look at the constructor of the class. Several properties are declared.

Properties can be set in the configuration file.

Change the configuration of the JetClustering algorithm in the following way:

genjet_clustering = JetClustering_MCParticleCollection_GenJetCollection_(
    "GenJetClustering",
    ptMin = 30.,
    )

And check the results.

3- Adding a filtering algorithm

The goal of this exercise is to fix the energy double counting obersved in exercise 1. The current algorithm sequence contains three algorithms: [reader, hepmc_converter, genjet_clustering] . We want to insert a new algorithm in the sequence, which would become: [reader, hepmc_converter, genp_filter, genjet_clustering] .

The new algorithm, genp_filter , will:

  • read in input the genparticles created by the hepmc_converter
  • write in output a copy of all the stable gen particles (status 1), discarding the other particles.

Go to the Generation/ directory.

cd Generation/src

Take the HepMCConverter as a base, and make a GenParticleFilter algorithm:

cp HepMCConverter.h GenParticleFilter.h
cp HepMCConverter.cpp GenParticleFilter.cpp

Since you created new files, next time you compile, you will need to do the following so that your files are detected by the build system:

cd $FCCSW
rm build.x86_64-slc6-gcc48-opt/CMakeCache.txt
make -j 4

When you just edit existing files, don't remove the CMakeCache.txt file, and just do:

cd $FCCSW
make -j 4

Edit the GenParticleFilter so that:

  • it takes in input a MCParticleCollection
  • it creates in output a MCParticleCollection (see HepMCConverter.cpp )
  • it copies only the particles with status 1 to the output collection (see the MCParticle.h and [ BareParticle.h classes to see which data members are available)

Finally, insert the new algorithm in the sequence in the configuration file, and run again. Check the printout and make sure that the jet energy is now what you would expect (around 45 GeV ).

-- ColinBernet {.twikiLink} - 11 December 2014