Introduction

This repository contains code for systematic testing and validation of the PPS direct simulation. It produces a set of standard plots (cf. the configs in templates directory) which can be compared across versions, settings etc. The tests are done in parallel for different PPS eras, cf. the settings directory.

Besides others, the standard plots include those quantifying the uncertainties related to the proton reconstruction: bias, resolution and systematics.

User's guide

  • Log on a computer of convenience (tested on CERN's LXPLUS).
  • Download this repository (e.g. with git clone).
  • Edit environment file: enter the path to your CMSSW installation (tested with CMSSW_12_1_0_pre3).
  • Initialise the environment: bash --rcfile environment.
  • Optionally: edit submit file and (un)comment the PPS eras for which you wish to run tests.
  • Produce execution scripts: ./submit <my_version>, where <my_version> stands for an output directory name.
  • If you wish to execute them on CERN's HTCondor, use the command printed on screen. If you wish to run (some of) them locally, use the run_multiple script - edit the file to select the version and the list of PPS eras. The local execution proceeds in background can take several minutes.
  • The results can be found in data/<my_version> directory.

Maintainer's guide

  • The scripts for evaluating reconstruction bias and resolution are in templates/proton_reco_resolution. Those for alignment and optics related systematics are in templates/proton_reco_misalignment and templates/proton_reco_optics, respectively.
  • The mis-alignment scenarios considered are listed in file templates/proton_reco_misalignment/run_multiple, around line 56, where additional details are given. Currently, 4 scenarios are considered, 2 in horizontal and 2 in vertical plane. For each projection, symmetric (near and far RPs are displaced in the same direction) and antisymmetric (opposite directions) are included. Should a new scenario be needed, another line can simply be added to the bash script.
  • The mis-alignment is simulated by using different geometries for simulation ("misaligned" geometry) and reconstruction ("real" geometry).
  • The list of optics-related systematic scenarios is give in file templates/proton_reco_optics/run_multiple around line 41, where further details are given.
  • The optics-related systematics are simulated by using different optics for simulation (default) and reconstruction (modified/biased). The modification/biasing of the optics is performed by the CMSSW module CTPPSModifiedOpticalFunctionsESSource. If new optics scenarios are added, or the existing ones are updated, the module must be updated first, then the list of scenarios shall be updated in the run_multiple script.
  • Whenever the list of systematics scenarios is changed, one needs to modify the file collect_systematics.cc (around line 136) and recompile it (run make).