Step 1 ntuples with Davinci

Info

DaVinci is the LHCb package that runs preliminary selections and calculations on the raw .dst data, and produces .root files. It is often run on LHC remote Linux nodes lxplus.

It is much more convenient to have a local DaVinci environment in a docker with a configuration that is easily shared. After the docker is pulled as described in the dependencies, it is launched from inside the project root with: 

make docker-dv

This command mounts the project root into the docker, so it has access to all the code in lhcb-ntuples-gen and allows you to modify it outside of the docker.

Launch DaVinci with additional TupleTool with this command, inside docker: 

run gaudirun.py <ntuple_options1> <ntuple_options2> ...
If you want to modify the TupleTools used in DaVinci, check the README on local development at https://github.com/umd-lhcb/TupleToolSemiLeptonic/tree/master.

We often have scripts that facilitate this. For instance, to run on Run 2 data, type in the docker: 

cd run2-rdx
./run.sh conds/cond-std-2016.py

Warning

The required .dst files need to be downloaded via git-annex OUTSIDE docker: 

cd run2-rdx
git annex get data/data-2016-md  # ~4.7 GB, this will take several minutes

Note

  • Inside the run script, the reco_Dst_D0.py script in the same directory is called for \(D^{*+}(\to D^0(\to K^+\pi^-)\pi^+)\mu^-\) reconstruction. It also sets how many events to run at most (EvtMax) and the print frequency (PrintFreq). To use a different reco script, edit run.sh.

  • The argument, conds/cond-std-2016.py, sets the type of input data (Data or MC), the input files and the name of the output file (std.root in this case).

Info

For more info about docker usage, refer to this guide.

For more info about Makefile, refer to this guide.