DataBase CellKiller
From ATLAS-TRIUMF
[edit] Introduction
Motivation:
Fake missing energy will be present in our data and not fully modeled in the MC (at least not at first)
- mis-modeled material
- High voltage problems
- Calo readout failures
Detector problems may be intermittent and therefore not detected by the calibrations. To prevent an analysis from mis-identifying events with fake MET as a signature for new physics, we need tools that can identify events with fake MET using variable from the event. The goal of this work is to develop methods and tools that can be implemented in a physics analysis to identify/correct/remove events with fake MET.
Database CellKiller Method:
In the original “CellKiller”, the energy in “unwanted” cells were simply set to zero during reconstruction. Now, in the “Database CellKiller”, cell killing is implemented using a database tag which contains the channels and FEBS that are to be turned off. During digitization, the Cell Killer database tag is read in and used to mask the affected cells. After digitization, the RDO files are reconstructed using the default job transform.
- In the “Database CellKiller”, the cells still exist but have no real energy. However, the cell still have noise, unlike the original “CellKiller”.
- In the “Database CellKiller”, the trigger information is valid.
[edit] MC Samples 15.3.1.5
The MC samples were produced in release 15.3.2.5 (for digitization and reconstruction) with the geometry ATLAS-GEO-08-00-00, conditionsTag OFLCOND-CSC-01-00-00 and the Database release is DBRelease-7.4.1.tar.gz. The trigger configuration used was MCRECO:MC_lumi1E31_no_prescale and the Jet reconstruction used RecJobTransforms/SetJetConstants-02-000.py.
Notes: The ESD and AOD are stored in the same output dataset. To run over the ESD add the command "--inputType=EXT0" to your pathena command. To run over AOD add the command "--inputType=EXT1" to your pathena command.
MC samples produced with the dead FEBs masked only during the reconstruction stage and the Tile failures in LBA22 and LBC46 (default reconstruction)
user10.IanMNugent.105010.J1_pythia_jetjet_84LArBadplusTileBad_digimaskon_2DCellCorron user10.IanMNugent.105011.J2_pythia_jetjet_84LArBadplusTileBad_digimaskon_2DCellCorron user10.IanMNugent.105012.J3_pythia_jetjet_84LArBadplusTileBad_digimaskon_2DCellCorron user10.IanMNugent.105013.J4_pythia_jetjet_84LArBadplusTileBad_digimaskon_2DCellCorron user10.IanMNugent.105014.J5_pythia_jetjet_84LArBadplusTileBad_digimaskon_2DCellCorron user10.IanMNugent.mc09_valid.108001.PythiaPhotonJet1_84LArBadplusTileBad_digimaskon_2DCellCorron user10.IanMNugent.mc09_valid.108002.PythiaPhotonJet2_84LArBadplusTileBad_digimaskon_2DCellCorron user10.IanMNugent.mc09_valid.108003.PythiaPhotonJet3_84LArBadplusTileBad_digimaskon_2DCellCorron user10.IanMNugent.mc09_valid.108004.PythiaPhotonJet4_84LArBadplusTileBad_digimaskon_2DCellCorron
MC samples produced with the dead FEBs masked only during the reconstruction stage (default reconstruction)
user09.IanMNugent.105010.J1_pythia_jetjet_84LArBad_2DCellCorron_DigiMaskInReco user09.IanMNugent.105011.J2_pythia_jetjet_84LArBad_2DCellCorron_DigiMaskInReco user09.IanMNugent.105012.J3_pythia_jetjet_84LArBad_2DCellCorron_DigiMaskInReco user09.IanMNugent.105013.J4_pythia_jetjet_84LArBad_2DCellCorron_DigiMaskInReco user09.IanMNugent.105014.J5_pythia_jetjet_84LArBad_2DCellCorron_DigiMaskInReco
MC samples produced with the Database CellKiller applied to simulate the 84 dead FEBS with the default reconstruction:
user09.IanMNugent.105010.J1_pythia_jetjet_84LArBad_digimaskon_2DCellCorron user09.IanMNugent.105011.J2_pythia_jetjet_84LArBad_digimaskon_2DCellCorron user09.IanMNugent.105012.J3_pythia_jetjet_84LArBad_digimaskon_2DCellCorron user09.IanMNugent.105013.J4_pythia_jetjet_84LArBad_digimaskon_2DCellCorron user09.IanMNugent.105014.J5_pythia_jetjet_84LArBad_digimaskon_2DCellCorron
MC samples produced with the Database CellKiller applied to simulate the 84 dead FEBS with the no correction from CaloCellNeighborsAverageCorr in reconstruction:
user09.IanMNugent.105010.J1_pythia_jetjet_84LArBad_digimaskon_NOCellCorron user09.IanMNugent.105011.J2_pythia_jetjet_84LArBad_digimaskon_NOCellCorron user09.IanMNugent.105012.J3_pythia_jetjet_84LArBad_digimaskon_NOCellCorron user09.IanMNugent.105013.J4_pythia_jetjet_84LArBad_digimaskon_NOCellCorron user09.IanMNugent.105014.J5_pythia_jetjet_84LArBad_digimaskon_NOCellCorron
MC samples produced with the Database CellKiller applied to simulate the 84 dead FEBS with the 3D correction from CaloCellNeighborsAverageCorr in reconstruction:
user09.IanMNugent.105010.J1_pythia_jetjet_84LArBad_digimaskon_3DCellCorron user09.IanMNugent.105011.J2_pythia_jetjet_84LArBad_digimaskon_3DCellCorron user09.IanMNugent.105012.J3_pythia_jetjet_84LArBad_digimaskon_3DCellCorron user09.IanMNugent.105013.J4_pythia_jetjet_84LArBad_digimaskon_3DCellCorron user09.IanMNugent.105014.J5_pythia_jetjet_84LArBad_digimaskon_3DCellCorron
MC samples produced with the Database CellKiller applied to simulate the 84 dead FEBS with the SUPER3D correction from CaloCellNeighborsAverageCorr in reconstruction:
user09.IanMNugent.105010.J1_pythia_jetjet_84LArBad_digimaskon_SUPER3DCellCorron user09.IanMNugent.105011.J2_pythia_jetjet_84LArBad_digimaskon_SUPER3DCellCorron user09.IanMNugent.105012.J3_pythia_jetjet_84LArBad_digimaskon_SUPER3DCellCorron user09.IanMNugent.105013.J4_pythia_jetjet_84LArBad_digimaskon_SUPER3DCellCorron user09.IanMNugent.105014.J5_pythia_jetjet_84LArBad_digimaskon_SUPER3DCellCorron
MC samples produced with no dead FEBs and default reconstruction (control samples)
user09.IanMNugent.105010.J1_pythia_jetjet_84LArBad_NOdigimask_2DCellCorron user09.IanMNugent.105011.J2_pythia_jetjet_84LArBad_NOdigimask_2DCellCorron user09.IanMNugent.105012.J3_pythia_jetjet_84LArBad_NOdigimask_2DCellCorron user09.IanMNugent.105013.J4_pythia_jetjet_84LArBad_NOdigimask_2DCellCorron user09.IanMNugent.105014.J5_pythia_jetjet_84LArBad_NOdigimask_2DCellCorron
Dead Regions
These MC sample have the 84 FEBs known to have OTx related problems simulated as dead to emulate the worst case scenario for the OTx optical read out failures. Below is a figure illustrating the dead regions.
Dead FEBs (17) Barrel=0/End=1 Cap Side FT SLOT 0 0 1 2 0 0 2 12 0 0 12 3 0 0 23 2 0 0 0 8 0 1 2 10 0 1 6 3 0 1 12 14 1 0 5 11 1 0 5 8 1 0 17 7 1 0 21 14 1 1 9 8 1 1 20 12 1 1 13 9 1 1 24 8 1 1 17 7 Switched off FEBs (67) Barrel=0/End=1 Cap Side FT SLOT 0 1 0 7 0 1 0 11 0 1 31 2 0 1 4 5 0 1 15 9 0 1 15 7 0 1 15 1 0 1 18 7 0 1 18 12 0 1 17 12 0 1 19 14 0 1 22 9 0 1 21 8 0 1 24 8 0 1 26 9 0 1 28 2 0 1 28 6 0 1 30 9 0 0 14 1 0 0 12 12 0 0 11 4 0 0 9 12 0 0 5 6 0 0 0 7 0 0 31 2 0 0 30 13 0 0 27 2 0 0 24 6 0 0 23 1 0 0 22 11 0 0 22 12 0 0 21 11 0 0 18 8 0 0 17 7 1 1 0 4 1 1 5 9 1 1 5 1 1 1 4 6 1 1 8 10 1 1 12 10 1 1 12 13 1 1 11 8 1 1 18 9 1 1 18 4 1 1 17 9 1 1 17 5 1 1 20 3 1 1 9 5 1 0 12 7 1 0 12 10 1 0 11 4 1 0 5 6 1 0 4 13 1 0 1 5 1 0 0 3 1 0 0 5 1 0 0 7 1 0 20 1 1 0 18 9 1 0 18 7 1 0 14 2 1 0 14 6 1 0 9 1 1 1 3 8 1 1 22 8 1 0 6 2 1 0 20 10
[edit] MC Samples 14.5.2.3
MC samples with the Database CellKiller applied:
AOD produced using release 14.5.2.3 for digitization and reconstruction
user09.IanMNugent.mc08.108001.PythiaPhotonJet1.simul.HITS.e344_s479.reco.ck_EXT1 user09.IanMNugent.mc08.108002.PythiaPhotonJet2.simul.HITS.e344_s479.reco.ck_EXT1 user09.IanMNugent.mc08.108003.PythiaPhotonJet3.simul.HITS.e344_s479.reco.ck_EXT1 user09.IanMNugent.mc08.108004.PythiaPhotonJet4.simul.HITS.e344_s479.reco.ck_EXT1 user09.IanMNugent.mc08.105010.J1_pythia_jetjet.simul.HITS.e344_s475.reco.ck_EXT1 user09.IanMNugent.mc08.105011.J2_pythia_jetjet.simul.HITS.e344_s479_tid026865.reco.ck_EXT1 user09.IanMNugent.mc08.105012.J3_pythia_jetjet.simul.HITS.e344_s479_tid026866.reco.ck_EXT1 user09.IanMNugent.mc08.105013.J4_pythia_jetjet.simul.HITS.e344_s479_tid026867.reco.ck_EXT1 user09.IanMNugent.mc08.105014.J5_pythia_jetjet.simul.HITS.e344_s479_tid039687.reco.ck_EXT1 user09.IanMNugent.mc08.105015.J6_pythia_jetjet.simul.HITS.e344_s475.reco.ck_EXT1 user09.IanMNugent.mc08.105016.J7_pythia_jetjet.simul.HITS.e344_s479.reco.ck_EXT1 user09.LorraineCourneyea.mc08.105200.T1_McAtNlo_Jimmy.merge.HITS.e357_s462_s520_tid056879.reco.ck_EXT1 user09.IanMNugent.mc08.105403.SU3_jimmy_susy.simul.HITS.e352_s462.reco.ck_EXT1 user09.IanMNugent.mc08.105404.SU6_jimmy_susy.simul.HITS.e352_s462_tid024669.reco.ck_EXT1 user09.IanMNugent.mc08.106400.SU4_jimmy_susy.simul.HITS.e352_s462_tid024666.reco.ck.d_EXT1
ESD produced using release 14.5.2.3 for digitization and reconstruction
user09.IanMNugent.mc08.108001.PythiaPhotonJet1.simul.HITS.e344_s479.reco.ck_EXT0 user09.IanMNugent.mc08.108002.PythiaPhotonJet2.simul.HITS.e344_s479.reco.ck_EXT0 user09.IanMNugent.mc08.108003.PythiaPhotonJet3.simul.HITS.e344_s479.reco.ck_EXT0 user09.IanMNugent.mc08.108004.PythiaPhotonJet4.simul.HITS.e344_s479.reco.ck_EXT0 user09.IanMNugent.mc08.105010.J1_pythia_jetjet.simul.HITS.e344_s475.reco.ck_EXT0 user09.IanMNugent.mc08.105011.J2_pythia_jetjet.simul.HITS.e344_s479_tid026865.reco.ck_EXT0 user09.IanMNugent.mc08.105012.J3_pythia_jetjet.simul.HITS.e344_s479_tid026866.reco.ck_EXT0 user09.IanMNugent.mc08.105013.J4_pythia_jetjet.simul.HITS.e344_s479_tid026867.reco.ck_EXT0 user09.IanMNugent.mc08.105014.J5_pythia_jetjet.simul.HITS.e344_s479_tid039687.reco.ck_EXT0 user09.IanMNugent.mc08.105015.J6_pythia_jetjet.simul.HITS.e344_s475.reco.ck_EXT0 user09.IanMNugent.mc08.105016.J7_pythia_jetjet.simul.HITS.e344_s479.reco.ck_EXT0 user09.LorraineCourneyea.mc08.105200.T1_McAtNlo_Jimmy.merge.HITS.e357_s462_s520_tid056879.reco.ck_EXT0 user09.IanMNugent.mc08.105403.SU3_jimmy_susy.simul.HITS.e352_s462.reco.ck_EXT0 user09.IanMNugent.mc08.105404.SU6_jimmy_susy.simul.HITS.e352_s462_tid024669.reco.ck_EXT0 user09.IanMNugent.mc08.106400.SU4_jimmy_susy.simul.HITS.e352_s462_tid024666.reco.ck.d_EXT0
MC control samples (without the Database CellKiller applied):
AOD produced using release 14.5.2.3 for digitization and reconstruction
user09.IanMNugent.mc08.108001.PythiaPhotonJet1.simul.HITS.e344_s479.reco.oracle_EXT1 user09.IanMNugent.mc08.108002.PythiaPhotonJet2.simul.HITS.e344_s479.reco.oracle_EXT1 user09.IanMNugent.mc08.108003.PythiaPhotonJet3.simul.HITS.e344_s479.reco.oracle_EXT1 user09.IanMNugent.mc08.108004.PythiaPhotonJet4.simul.HITS.e344_s479.reco.oracle_EXT1 user09.IanMNugent.mc08.105010.J1_pythia_jetjet.simul.HITS.e344_s475.reco.oracle_EXT1 user09.IanMNugent.mc08.105011.J2_pythia_jetjet.simul.HITS.e344_s479_tid026865.reco.oracle_EXT1 user09.IanMNugent.mc08.105012.J3_pythia_jetjet.simul.HITS.e344_s479_tid026866.reco.oracle_EXT1 user09.IanMNugent.mc08.105013.J4_pythia_jetjet.simul.HITS.e344_s479_tid026867.reco.oracle_EXT1 user09.IanMNugent.mc08.105014.J5_pythia_jetjet.simul.HITS.e344_s479_tid039687.reco.oracle_EXT1 user09.IanMNugent.mc08.105015.J6_pythia_jetjet.simul.HITS.e344_s475.reco.oracle_EXT1 user09.IanMNugent.mc08.105016.J7_pythia_jetjet.simul.HITS.e344_s479.reco.oracle_EXT1 user09.LorraineCourneyea.mc08.105200.T1_McAtNlo_Jimmy.merge.HITS.e357_s462_s520_tid056879.reco.oracle_EXT1 user09.IanMNugent.mc08.105403.SU3_jimmy_susy.simul.HITS.e352_s462.reco.oracle_EXT1 user09.IanMNugent.mc08.105404.SU6_jimmy_susy.simul.HITS.e352_s462_tid024669.reco.oracle_EXT1 user09.IanMNugent.mc08.106400.SU4_jimmy_susy.simul.HITS.e352_s462_tid024666.reco.oracle.d_EXT1
ESD produced using release 14.5.2.3 for digitization and reconstruction
user09.IanMNugent.mc08.108001.PythiaPhotonJet1.simul.HITS.e344_s479.reco.oracle_EXT0 user09.IanMNugent.mc08.108002.PythiaPhotonJet2.simul.HITS.e344_s479.reco.oracle_EXT0 user09.IanMNugent.mc08.108003.PythiaPhotonJet3.simul.HITS.e344_s479.reco.oracle_EXT0 user09.IanMNugent.mc08.108004.PythiaPhotonJet4.simul.HITS.e344_s479.reco.oracle_EXT0 user09.IanMNugent.mc08.105010.J1_pythia_jetjet.simul.HITS.e344_s475.reco.oracle_EXT0 user09.IanMNugent.mc08.105011.J2_pythia_jetjet.simul.HITS.e344_s479_tid026865.reco.oracle_EXT0 user09.IanMNugent.mc08.105012.J3_pythia_jetjet.simul.HITS.e344_s479_tid026866.reco.oracle_EXT0 user09.IanMNugent.mc08.105013.J4_pythia_jetjet.simul.HITS.e344_s479_tid026867.reco.oracle_EXT0 user09.IanMNugent.mc08.105014.J5_pythia_jetjet.simul.HITS.e344_s479_tid039687.reco.oracle_EXT0 user09.IanMNugent.mc08.105015.J6_pythia_jetjet.simul.HITS.e344_s475.reco.oracle_EXT0 user09.IanMNugent.mc08.105016.J7_pythia_jetjet.simul.HITS.e344_s479.reco.oracle_EXT0 user09.LorraineCourneyea.mc08.105200.T1_McAtNlo_Jimmy.merge.HITS.e357_s462_s520_tid056879.reco.oracle_EXT0 user09.IanMNugent.mc08.105403.SU3_jimmy_susy.simul.HITS.e352_s462.reco.oracle_EXT0 user09.IanMNugent.mc08.105404.SU6_jimmy_susy.simul.HITS.e352_s462_tid024669.reco.oracle_EXT0 user09.IanMNugent.mc08.106400.SU4_jimmy_susy.simul.HITS.e352_s462_tid024666.reco.oracle.d_EXT0
Dead regions:
Barrel/End Cap Side FT SLOT Barrel 0 10 1-14 Barrel 0 11 1-14 End Cap 1 2 1-14 End Cap 1 3 1-14
Below are Plot of the number of CaloClusters (Figures 1-5) and Cone4H1TowerJets (Figures 6-10) as a function of eta and phi which demonstrate the Database CellKiller Software.
Figure 1: The number of CaloClusters in the EMB as a function of eta and phi for the J5 MC sample with the Database CellKiller applied.
Figure 2: The number of CaloClusters in the EME as a function of eta and phi for the J5 MC sample with the Database CellKiller applied.
Figure 3: The number of CaloClusters in the HEC as a function of eta and phi for the J5 MC sample with the Database CellKiller applied.
Figure 4: The number of CaloClusters in the FCal as a function of eta and phi for the J5 MC sample with the Database CellKiller applied.
Figure 5: The number of CaloClusters in the Tile as a function of eta and phi for the J5 MC sample with the Database CellKiller applied.
Figure 6: The number of Cone4H1TowerJets in the EMB as a function of eta and phi for the J5 MC sample with the Database CellKiller applied.
Figure 7: The number of Cone4H1TowerJets in the EME as a function of eta and phi for the J5 MC sample with the Database CellKiller applied.
Figure 8: The number of Cone4H1TowerJets in the HEC as a function of eta and phi for the J5 MC sample with the Database CellKiller applied.
Figure 9: The number of Cone4H1TowerJets in the FCal as a function of eta and phi for the J5 MC sample with the Database CellKiller applied.
Figure 10: The number of Cone4H1TowerJets in the Tile as a function of eta and phi for the J5 MC sample with the Database CellKiller applied.












