Add `SimBeamSpotObjects` to MC Global Tags

[francescobrivio]

Multi-step solution for cms-sw/cmssw#41894:

• Update SimBeamSpotObjects code
  • Done in Update utilities for SimBeamSpotObjects handling cms-sw/cmssw#41918 and Miscellaneous improvements to SimBeamSpotObjects handling utilities cms-sw/cmssw#41967
• Make payloads and tags for the new objects
  • Create the payloads
    • Already created by @mmusich (thanks a lot!) and upload to the Prep DB here
  • Upload them to Prod CondDB
  • Announce them on CMSTalk (done here)
• Include the new tags in the MC GTs
  • Need to check:
    • Which GTs should actually be updated, only Run3 or also Run2 and Run1? done here
    • The correct correspondence between GT <-> tag done here
• PPS experts will have to adapt the code (done here)
• Eventually the central CMS MC productions should be moved to consume the VtxSmearing from the GT
  • This is probably already implemented, but I need to find where exactly in the code done in Read VertexSmearing from GT in Run-1/2/3 MC cms-sw/cmssw#43242

[mmusich]

In order to help with this step

• Include the new tags in the MC GTs
  • Need to check:
    • Which GTs should actually be updated, only Run3 or also Run2 and Run1?
    • The correct correspondence between GT <-> tag

and hence make progress with cms-sw/cmssw#42664 I started to compile the coupling Global Tag <-> SimBeamSpot tag for each simulation GT key of autoCond (as of CMSSW_13_3_X_2023-08-25-2300), in order to decide which tags to move from the preparation to the production database.
You can find it below:

Follows list of GT ⇆ tag couplings

• run1_design : 131X_mcRun1_design_v2

  • BeamSpotObjectsRcd : Realistic8TeVCollisions_START50_V13_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic8TeVCollisionVtxSmearingParameters_v0_mc,

• run1_mc : 131X_mcRun1_realistic_v2

  • BeamSpotObjectsRcd : Realistic8TeVCollisions_START50_V13_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic8TeVCollisionVtxSmearingParameters_v0_mc

• run1_mc_hi : 131X_mcRun1_HeavyIon_v2

  • BeamSpotObjectsRcd : RealisticHICollisions2013_START53_V9_v1_mc
    *SimBeamSpotObjectsRcd : SimBeamSpot_RealisticHIpPb2013CollisionVtxSmearingParameters_v0_mc,

• run2_mc_50ns : 131X_mcRun2_startup_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic50ns_13TeVCollisions_Startup_v0_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic50ns13TeVCollisionVtxSmearingParameters_v0_mc,

• run2_mc_l1stage1 : 131X_mcRun2_asymptotic_l1stage1_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic50ns_13TeVCollisions_Asymptotic_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic50ns13TeVCollisionVtxSmearingParameters_v0_mc,

• run2_mc_pre_vfp : 131X_mcRun2_asymptotic_preVFP_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13TeV2016Collisions_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13TeV2016CollisionVtxSmearingParameters_v0_mc,

• run2_mc : 131X_mcRun2_asymptotic_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13TeV2016Collisions_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13TeV2016CollisionVtxSmearingParameters_v0_mc,

• run2_mc_cosmics : 131X_mcRun2cosmics_asymptotic_deco_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13TeV2016Collisions_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13TeV2016CollisionVtxSmearingParameters_v0_mc,

• run2_mc_hi : 131X_mcRun2_HeavyIon_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_RealisticHICollision2015_MC_75X_v1
  • SimBeamSpotObjectsRcd : SimBeamSpot_RealisticHICollisionFixZ2015VtxSmearingParameters_v0_mc,

• run2_mc_pa : 131X_mcRun2_pA_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic8TeVPACollision2016_v1
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic8TeVPACollision2016VtxSmearingParameters_v0_mc,

• phase1_2017_realistic : 131X_mc2017_realistic_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13TeVCollisions_Early2017_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13TeVEarly2017CollisionVtxSmearingParameters_v0_mc,

• phase1_2017_cosmics : 131X_mc2017cosmics_realistic_deco_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13TeVCollisions_Early2017_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13TeVEarly2017CollisionVtxSmearingParameters_v0_mc,

• phase1_2017_cosmics_peak : 131X_mc2017cosmics_realistic_peak_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13TeVCollisions_Early2017_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13TeVEarly2017CollisionVtxSmearingParameters_v0_mc,

• phase1_2018_realistic : 131X_upgrade2018_realistic_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13TeVCollisions_Early2018_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13TeVEarly2018CollisionVtxSmearingParameters_v0_mc,

• phase1_2018_realistic_rd : 131X_upgrade2018_realistic_RD_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13TeVCollisions_Early2018_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13TeVEarly2018CollisionVtxSmearingParameters_v0_mc,

• phase1_2018_realistic_hi : 131X_upgrade2018_realistic_HI_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic_5TeVPbPb_HI2018A_MC_v1
  • SimBeamSpotObjectsRcd : SimBeamSpot_RealisticPbPbCollision2018VtxSmearingParameters_v0_mc

• phase1_2018_realistic_HEfail : 131X_upgrade2018_realistic_HEfail_v3

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13TeVCollisions_Early2018_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13TeVEarly2018CollisionVtxSmearingParameters_v0_mc,

• phase1_2018_cosmics : 131X_upgrade2018cosmics_realistic_deco_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13TeVCollisions_Early2018_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13TeVEarly2018CollisionVtxSmearingParameters_v0_mc,

• phase1_2018_cosmics_peak : 131X_upgrade2018cosmics_realistic_peak_v3

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13TeVCollisions_Early2018_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13TeVEarly2018CollisionVtxSmearingParameters_v0_mc,

• phase1_2022_realistic : 132X_mcRun3_2022_realistic_v1

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13p6TeVCollisions_EOY2022_v2_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13p6TeVEOY2022CollisionVtxSmearingParameters_v0_mc,

• phase1_2022_realistic_postEE : 132X_mcRun3_2022_realistic_postEE_v1

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13p6TeVCollisions_EOY2022_v2_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13p6TeVEOY2022CollisionVtxSmearingParameters_v0_mc,

• phase1_2022_cosmics : 132X_mcRun3_2022cosmics_realistic_deco_v1

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13p6TeVCollisions_EOY2022_v2_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13p6TeVEOY2022CollisionVtxSmearingParameters_v0_mc,

• phase1_2022_realistic_hi : 132X_mcRun3_2022_realistic_HI_v1

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic2022PbPbCollision_v5_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic2022PbPbCollisionVtxSmearingParameters_v0_mc,

• phase1_2023_realistic : 132X_mcRun3_2023_realistic_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13p6TeVCollisions_Early2023_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13p6TeVEarly2023CollisionVtxSmearingParameters_v0_mc,

• phase1_2023_realistic_postBPix : 132X_mcRun3_2023_realistic_postBPix_v1

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13p6TeVCollisions_Early2023_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13p6TeVEarly2023CollisionVtxSmearingParameters_v0_mc,

• phase1_2023_cosmics : 132X_mcRun3_2023cosmics_realistic_deco_v2

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13p6TeVCollisions_Early2023_v1_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13p6TeVEarly2023CollisionVtxSmearingParameters_v0_mc,

• phase1_2023_realistic_hi : 132X_mcRun3_2023_realistic_HI_v1

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic2022PbPbCollision_v5_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic2022PbPbCollisionVtxSmearingParameters_v0_mc,

• phase1_2024_realistic : 132X_mcRun3_2024_realistic_v1

  • BeamSpotObjectsRcd : BeamSpotObjects_Realistic25ns_13p6TeVCollisions_Early2022_v3_mc
  • SimBeamSpotObjectsRcd : SimBeamSpot_Realistic25ns13p6TeVEarly2022CollisionVtxSmearingParameters_v0_mc,

[mmusich]

While compiling this list, one issue with this approach that occurred to me, is that the current SimBeamSpotObjects CondFormat assumes that the simulation step happens via the BetafuncEvtVtxGenerator, while there are instead use cases that use a different functional form (e.g. in the case of the design simulation the GaussEvtVtxGenerator or in the case of phase-2 simulation the HLLHCEvtVtxGenerator).
These use a different set of parameters that cannot be translated into the existing SimBeamSpotObjects data layout.
I guess it's up to discussion if (given that currently there is no tag associated to this C++ class in the production database) if it's appropriate to amend the class layout to allow these parameters, or if otherwise devise entirely different classes (and hence EventSetup records).

Below the list of PSets in IOMC/EventVertexGenerators/python/VtxSmearedParameters_cfi.py affected by this problem:

GaussVtxSigmaZ4cmSmearingParameters : cms.PSet(
    MeanX = cms.double(0.0),
    MeanY = cms.double(0.0),
    MeanZ = cms.double(0.0),
    SigmaX = cms.double(0.0015),
    SigmaY = cms.double(0.0015),
    SigmaZ = cms.double(4.0),
    TimeOffset = cms.double(0.0)
)
GaussVtxSmearingParameters : cms.PSet(
    MeanX = cms.double(0.0),
    MeanY = cms.double(0.0),
    MeanZ = cms.double(0.0),
    SigmaX = cms.double(0.0015),
    SigmaY = cms.double(0.0015),
    SigmaZ = cms.double(5.3),
    TimeOffset = cms.double(0.0)
)
HLLHCCrabKissingVtxSmearingParameters : cms.PSet(
    BeamProfile = cms.string('Flat'),
    BetaStarCrossingPlaneInm = cms.double(0.3),
    BetaStarParallelPlaneInm = cms.double(0.075),
    CrabAngleCrossingPlaneInurad = cms.double(200.0),
    CrabAngleParallelPlaneInurad = cms.double(100.0),
    CrabFrequencyCrossingPlaneInMHz = cms.double(400.0),
    CrabFrequencyParallelPlaneInMHz = cms.double(400.0),
    EprotonInGeV = cms.double(6500.0),
    HalfCrossingAngleInurad = cms.double(200.0),
    MeanXIncm = cms.double(0.0),
    MeanYIncm = cms.double(0.0),
    MeanZIncm = cms.double(0.0),
    NormalizedEmittanceCrossingPlaneInum = cms.double(2.5),
    NormalizedEmittanceParallelPlaneInum = cms.double(2.5),
    TimeOffsetInns = cms.double(0.0),
    ZsizeInm = cms.double(0.15)
)
HLLHCVtxSmearingParameters : cms.PSet(
    BetaCrossingPlaneInm = cms.double(0.2),
    BetaSeparationPlaneInm = cms.double(0.2),
    BunchLengthInm = cms.double(0.09),
    CrabFrequencyInMHz = cms.double(400.0),
    CrabbingAngleCrossingInurad = cms.double(380.0),
    CrabbingAngleSeparationInurad = cms.double(0.0),
    CrossingAngleInurad = cms.double(510.0),
    EprotonInGeV = cms.double(6500.0),
    HorizontalEmittance = cms.double(2.5e-06),
    MeanXIncm = cms.double(0.0),
    MeanYIncm = cms.double(0.0),
    MeanZIncm = cms.double(0.0),
    RF800 = cms.bool(False),
    TimeOffsetInns = cms.double(0.0),
    VerticalEmittance = cms.double(2.05e-06)
)
Run3FlatOpticsGaussVtxSigmaZ4p2cmSmearingParameters : cms.PSet(
    MeanX = cms.double(0.0107682),
    MeanY = cms.double(0.041722),
    MeanZ = cms.double(0.035748),
    SigmaX = cms.double(0.00118),
    SigmaY = cms.double(0.00055),
    SigmaZ = cms.double(4.2),
    TimeOffset = cms.double(0.0)
)
Run3FlatOpticsGaussVtxSigmaZ5p3cmSmearingParameters : cms.PSet(
    MeanX = cms.double(0.0107682),
    MeanY = cms.double(0.041722),
    MeanZ = cms.double(0.035748),
    SigmaX = cms.double(0.0015),
    SigmaY = cms.double(0.0013),
    SigmaZ = cms.double(5.3),
    TimeOffset = cms.double(0.0)
)

[civanch]

@mmusich , would be correct if we will try to resolve the issue in the next release cycle? The dead-line for 13_3 is too close.

[mmusich]

@civanch

would be correct if we will try to resolve the issue in the next release cycle? The dead-line for 13_3 is too close.

Personally, I don't mind. I am waiting the alca people to comment on the points above. @saumyaphor4252 @francescobrivio

[mmusich]

may I ask where we stand with this issue . @saumyaphor4252 @francescobrivio

[francescobrivio]

I plan to take a look this afternoon, sorry for the delay!

[francescobrivio]

Thanks @mmusich for compiling the list and pointing out the missing ingredients!
Regarding:

if it's appropriate to amend the class layout to allow these parameters, or if otherwise devise entirely different classes (and hence EventSetup records).

I compiled a list of all the members that would be needed:

fX0          // for BetafuncEvtVtxGenerator only
fY0
fZ0
fbetastar
femittance
fPhi
fAlpha

fSigmaZ      // for BetafuncEvtVtxGenerator and GaussEvtVtxGenerator

fSigmaX      // for GaussEvtVtxGenerator only
fSigmaY

fMeanX       // for GaussEvtVtxGenerator and HLLHCEvtVtxGenerator
fMeanY
fMeanZ

momeV        // for HLLHCEvtVtxGenerator only
phi
wcc
RF800
betx
bets
epsxn
epssn
sigs
alphax
alphay

fTimeOffset  // for all three generators

So, given that the original SimBeamSpotObjects alraedy includes 9 of them, we can update it to include them all.

But there are a few parameters defined in HLLHCCrabKissingVtxSmearingParameters:

     BeamProfile = cms.string('Flat'),
     BetaStarCrossingPlaneInm = cms.double(0.3),
     BetaStarParallelPlaneInm = cms.double(0.075),
     CrabAngleCrossingPlaneInurad = cms.double(200.0),
     CrabAngleParallelPlaneInurad = cms.double(100.0),
     CrabFrequencyCrossingPlaneInMHz = cms.double(400.0),
     CrabFrequencyParallelPlaneInMHz = cms.double(400.0),
     HalfCrossingAngleInurad = cms.double(200.0),
     NormalizedEmittanceCrossingPlaneInum = cms.double(2.5),
     NormalizedEmittanceParallelPlaneInum = cms.double(2.5),
     ZsizeInm = cms.double(0.15)

That are not used in any .cc/.h file (as a matter of fact they are not used anywhere as this example search in cmssw shows), so I'm not sure how to treat those...

Alternatively one could think about updating the current simBS object with the parameters needed by the GaussEvtVtxGenerator now, and postpone this migration for the Phase2 GT to later, by making an entirely new record/object for the HLLHC case. Would this be technically doable (i.e. to decaouple the Phase2 GT update)?

What do you think? I'm open to suggestions!

[mmusich]

Alternatively one could think about updating the current simBS object with the parameters needed by the GaussEvtVtxGenerator now, and postpone this migration for the Phase2 GT to later, by making an entirely new record/object for the HLLHC case. Would this be technically doable (i.e. to decaouple the Phase2 GT update)?

I think it's probably better to settle the question now that the record is "virgin" (never used in production). About the parameters never referenced in the source code we should follow up with the developers that created the configuration in the first place.

[francescobrivio]

Dear @lgray,
sorry to drag you into this discussion, but from github history it seems like you were the last one to update the HL-LHC beamspot parameters in CMSSW!

We are trying to migrate the vertex smearing parameters to the GT (see issue description and the following discussion for details), and we noticed that in cms-sw/cmssw#16942 you removed some of the parameters used by IOMC/EventVertexGenerators/interface/HLLHCEvtVtxGenerator.h, but the same parameters were never cleaned up in the HLLHCCrabKissingVtxSmearingParameters configuration, see:
https://github.com/cms-sw/cmssw/blob/2cd78edfb8f8a002926332e7cd5eedbd000501b6/IOMC/EventVertexGenerators/python/VtxSmearedParameters_cfi.py#L1080-L1098

Was that done on purpose? Or simply forgotten?

Thanks a lot!

Cheers,
Francesco

[francescobrivio]

First part of this issue, regarding Run-1/2/3 GTs, is almost done, see:
https://cms-talk.web.cern.ch/t/mc-new-simbeamspotobjectsrcd-tags-for-run1-2-3-mc-gts/31292

For the Phase-2 GT: the work to define a new CondFormat is ongoing in cms-sw/cmssw#43186