MDC2020: Difference between revisions

From Mu2eWiki
Jump to navigation Jump to search
Line 166: Line 166:


The digitization stream is divided into 'Triggered', ie events which passed at least one reco-object based trigger selections, and 'Triggerable', ie events which pass a MC-truth based filter similar to the trigger selection.  Note that these sets are not miscible (cannot be read in the same job) and the events selected do not have any pre-determined relation, ie one is not a subset of the other.
The digitization stream is divided into 'Triggered', ie events which passed at least one reco-object based trigger selections, and 'Triggerable', ie events which pass a MC-truth based filter similar to the trigger selection.  Note that these sets are not miscible (cannot be read in the same job) and the events selected do not have any pre-determined relation, ie one is not a subset of the other.


Current datasets:
Current datasets:

Revision as of 15:57, 3 December 2024

Introduction

MDC2020 is an end-to-end production using updated geometry, detector simulations, persistent schema, and simulation workflows compared to MDC2018. The goal is to provide a reasonably complete and accurate model of what Mu2e will record during commissioning and first running period ('Run 1', add reference), including OnSpill, OffSpill, and Extracted Position samples. The primary intended use case of these samples are:

  • General Notes and Updates
  • Detector calibration and alignment, including cross-system (ie CRV to Calo) calibrations
  • Detector commissioning
  • Trigger algorithm testing and development
  • TDAQ to Offline data transfer workflows
  • Offline reconstruction algorithm development
  • Science extraction framework and algorithm development

The detailed technical documentation of MDC2020 is in Doc-DB 44084. These pages provide a list of the datasets produced and give time-dependent information.

General Notes and Updates

As reported in this presentation the G4 code used in all the MDC2020 simulations had missing contributions from delta rays in the calorimeter. This appears to mostly affect signals from cosmic muons. For precision studies using cosmic muons please contact the experts.

Simulation Workflow

The samples described on this page were produced using the following workflows. Jobs were submitted using POMS.

The workflow used for MDC2020.

Cosmic Simulation campaigns

The cosmic simulation creates a sample of cosmic rays with three different configurations:

  • Standard position, magnetic field on
  • Standard position, magnetic field off
  • Extracted position, magnetic field off


Standard position, magnetic field on - CORSIKA

Stage S1 Size [T] Event count [1M] Event count/per file Livetime [s] per file nFiles Version Live-time [sec] Resampling factor
S1 sim.mu2e.CosmicDSStopsCORSIKA.MDC2020ab.art 0.122 1.8E+07 215.1 0.0534 82,330 MDC2020ab 4394 NA
S1 sim.mu2e.CosmicDSStopsCORSIKALow.MDC2020ab.art 0.019 2.7E+06 32.6 0.0534 82,332 MDC2020ab 4394 NA
S2 dts.mu2e.CosmicCORSIKALow.MDC2020ae.art 30 0.6E+09 7372 850 78,942 MDC2020ae 6.7E7 15000
S2 dts.mu2e.CosmicCORSIKAAll.MDC2020ae.art 36 4.3E+09 53346 123 81,134 MDC2020ae 1.0E+07 2,278
S2 filt dts.mu2e.CosmicCORSIKASignalAll.MDC2020ae.art 0.33 9.6E+06 11792 12469 812 MDC2020ae 1.0E+07 2,278
S2 filt dts.mu2e.CosmicCORSIKACalibAll.MDC2020ae.art 0.10 1.1E+07 13833 125 812 MDC2020ae 1.0E+05
S2 filt dts.mu2e.CosmicCORSIKASignalAll.MDC2020ag.art 0.33 9.6E+06 11792 12469 81,134 MDC2020ag 1.0E+07 2,278
S2 filt dts.mu2e.CosmicCORSIKACalibAll.MDC2020ag.art 0.10 1.1E+07 13833 125 81,134 MDC2020ag 1.0E+05

Standard position, magnetic field on - CRY

Stage S1 Size [T] Event count Event count/per file Livetime [s] per file nFiles Version Live-time [sec] Resampling factor
S1 sim.mu2e.CosmicDSStopsCRY.010622.art 13.6 2.0E+09 42280.9 11.835 46,981 MDC2020n 556000 NA
S1 sim.mu2e.CosmicDSStopsLowCRY.010622.art 0.71 2.4E+08 5204.2 11.835 46,981 MDC2020n 556000 NA
S2 dts.mu2e.CosmicCRYAll.MDC2020ae.art 39 4.4E+09 54007 130 82,141 MDC2020ae 1.1E+07 19
S2 filt dts.mu2e.CosmicCRYSignalAll.MDC2020ae.art 0.36 1E7 12226 12991 822 MDC2020ae 1.1E+07 19
S2 filt dts.mu2e.CosmicCRYCalibAll.MDC2020ae.art 0.11 1.2E7 14150 130 822 MDC2020ae 1.1E+05

Beam Simulation POMS campaign

The goal of the beam campaign is to generate the particles needed for mixing (muons, electrons, neutrals) starting from the protons on target (POT stage).

Digitization and mixing POMS campaign

POMS campaign for primary particle.

For each primary (e.g. DIO, CeEndpoint, CePlusEndpoint, etc.) we run digitization and reconstruction with ("mix" output) and without beam mixing ("digi" output). The beam mixing stream is divided into three branches one per configuration ("perfect", "best", "reco").

To run mixing on a new primary you should use a clone of the production POMS campaign, substituting your primary for one of the existing (CeEndpont, etc). Note that the new primary must have been entered into the SAM database as part of MDC2020 production, preferably using a clone of the primary POMS campaign.

to TEST mixing, you can manually invoke the scripts, as in the following example:

> cd mymusedir
> muse setup
> setup dhtools
> setup mu2etools
> kx509
> vomsCert
> git clone git@github.com:Mu2e/Production.git
> source Production/Scripts/gen_Mix.sh CeEndpoint MDC2020 p r v perfect v2_0 1BB
> mu2e -c CeEndpointMix1BB_000/cnf.mu2e.CeEndpointMix1BB.MDC2020v_perfect_v2_0.001210_00000000.fcl --nevts 10

Recent Campaigns

The most recent MDC2020 campaigns are detailed below along with their Offline and Production versions as well as any useful information:

Campaign Offline ver Production ver mu2e_trig_config ver Comments
MDC2020ae v10_29_00 v00_21_00 v01_02_00 Re-digitization and re-reconstruction of all datasets, new CORSIKA datasets
MDC2020z v10_23_01 v00_16_00 - Re-reconstruction of some datasets
MDC2020v v10_20_00 v00_12_00 - Special run (see below), not for standard physics studies!
MDC2020t v10_17_00 v00_09_11 - several updates to many directories
MDC2020r v00_09_02 v10_15_01 - bug fix for digi stage

Current Datasets

These are the datasets currently available which correspond to the full 100% campaign. The three digi and reco configurations ("perfect", "best", "reco") correspond to three different detector conditions as described in DocDB 42036.

To obtain details (number of events, number of files, GB, etc.) about a dataset you can use the datasetSummary.sh script, available in the Production repository.

Note that, to obtain consistent results, you should use the same database version to process a dataset as it was produced with. Thus to read dig.mu2e.CeEndpointOnSpillTriggered.MDC2020ae_best_v1_3.art, you should configure your job to use database purpose 'best', version v1_3. For more details see the description of the conditions database.

Note that datasets can require specific versions of our codes to process, as documented in the following table. Use of an inconsistent code version may result in job failure (exception) or incorrect results.

Database version of dataset Offline tag for processing dig (digis) TrkAna tag for processing mcs (reco)
v1_3.art >= v10_29_00 or HEAD >= v05_02_00 or HEAD
v1_1.art <= v10_28_00 * <= v05_01_00
* Note that v1_1 and older digi datasets may still be processed by modern versions of Offline using additional fcl overrides described here. mcs files produced this way can be used with any version of TrkAna.

Reprocessed digitization and reconstruction datasets [2024]

In May 2024 several samples were remade from the dts->digi and digi-->reco stages to incorporate updates in digitization algorithms, triggering, trigger streams, reconstruction, and db changes.

The digitization stream is divided into 'Triggered', ie events which passed at least one reco-object based trigger selections, and 'Triggerable', ie events which pass a MC-truth based filter similar to the trigger selection. Note that these sets are not miscible (cannot be read in the same job) and the events selected do not have any pre-determined relation, ie one is not a subset of the other.

Current datasets: dig mcs

Mixed:

Digi Reco NTuple Comments
dig.mu2e.CeEndpointMix1BBTriggered.MDC2020ae_best_v1_3.art None
dig.mu2e.CeEndpointMix1BBTriggered.MDC2020ae_perfect_v1_3.art None
dig.mu2e.CeEndpointMix2BBTriggered.MDC2020ae_best_v1_3.art mcs.mu2e.CeEndpointMix2BBTriggered.MDC2020ae_best_v1_3.art
dig.mu2e.CeEndpointMix2BBTriggered.MDC2020ae_perfect_v1_3.art mcs.mu2e.CeEndpointMix2BBTriggered.MDC2020ae_perfect_v1_3.art
dig.mu2e.CePlusEndpointMix1BBTriggered.MDC2020ae_best_v1_3.art None
dig.mu2e.CePlusEndpointMix1BBTriggered.MDC2020ae_perfect_v1_3.art None
dig.mu2e.NoPrimaryMix1BBTriggered.MDC2020ae_best_v1_3.art mcs.mu2e.NoPrimaryMix1BBTriggered.MDC2020ae_best_v1_3.art
dig.mu2e.NoPrimaryMix1BBTriggered.MDC2020ae_perfect_v1_3.art mcs.mu2e.NoPrimaryMix1BBTriggered.MDC2020ae_perfect_v1_3.art
dig.mu2e.NoPrimaryMix2BBTriggered.MDC2020ae_best_v1_3.art mcs.mu2e.NoPrimaryMix2BBTriggered.MDC2020ae_best_v1_3.art
dig.mu2e.NoPrimaryMix2BBTriggered.MDC2020ae_perfect_v1_3.art mcs.mu2e.NoPrimaryMix2BBTriggered.MDC2020ae_perfect_v1_3.art


Primaries:

Digi Reco NTuple Comments
dig.mu2e.CeEndpointOnSpillTriggered.MDC2020ae_best_v1_3.art mcs.mu2e.CeEndpointOnSpillTriggered.MDC2020ae_best_v1_3.art
dig.mu2e.CeEndpointOnSpillTriggered.MDC2020ae_perfect_v1_3.art mcs.mu2e.CeEndpointOnSpillTriggered.MDC2020ae_perfect_v1_3.art
dig.mu2e.CePlusEndpointOnSpillTriggered.MDC2020ae_best_v1_3.art mcs.mu2e.CePlusEndpointOnSpillTriggered.MDC2020ae_best_v1_3.art
dig.mu2e.CePlusEndpointOnSpillTriggered.MDC2020ae_perfect_v1_3.art mcs.mu2e.CePlusEndpointOnSpillTriggered.MDC2020ae_perfect_v1_3.art
dig.mu2e.CosmicCORSIKACalibAllOffSpillTriggered.MDC2020ae_best_v1_3.art mcs.mu2e.CosmicCORSIKACalibAllOffSpillTriggered.MDC2020ae_best_v1_3.art
dig.mu2e.CosmicCORSIKACalibAllOffSpillTriggered.MDC2020ae_perfect_v1_3.art mcs.mu2e.CosmicCORSIKACalibAllOffSpillTriggered.MDC2020ae_perfect_v1_3.art
dig.mu2e.CosmicCORSIKACalibAllOnSpillTriggered.MDC2020ae_best_v1_3.art mcs.mu2e.CosmicCORSIKACalibAllOnSpillTriggered.MDC2020ae_best_v1_3.art
dig.mu2e.CosmicCORSIKACalibAllOnSpillTriggered.MDC2020ae_perfect_v1_3.art mcs.mu2e.CosmicCORSIKACalibAllOnSpillTriggered.MDC2020ae_perfect_v1_3.art
dig.mu2e.CosmicCORSIKASignalAllOffSpillTriggered.MDC2020ae_best_v1_3.art None
dig.mu2e.CosmicCORSIKASignalAllOffSpillTriggered.MDC2020ae_perfect_v1_3.art mcs.mu2e.CosmicCORSIKASignalAllOffSpillTriggered.MDC2020ae_perfect_v1_3.art
dig.mu2e.CosmicCORSIKASignalAllOnSpillTriggered.MDC2020ae_best_v1_3.art None
dig.mu2e.CosmicCORSIKASignalAllOnSpillTriggered.MDC2020ae_perfect_v1_3.art mcs.mu2e.CosmicCORSIKASignalAllOnSpillTriggered.MDC2020ae_perfect_v1_3.art
dig.mu2e.CosmicCRYExtractedCatTriggered.MDC2020ae_best_v1_3.art mcs.mu2e.CosmicCRYExtractedCatTriggered.MDC2020ae_best_v1_3.art

Reprocessed reconstruction datasets [2023]

In Sept 2023 several samples were remade from the digi-->reco stage to incorporate updates in reconstruction and db changes.

Digi Reco NTuple Comments
dig.mu2e.CeEndpointMix1BBSignal.MDC2020r_best_v1_0.art mcs.mu2e.CeEndpointMix1BBSignal.MDC2020z_best_v1_1.art nts.mu2e.CeEndpointMix1BBSignalMix1BB.MDC2020z1_best_v1_1_std_v04_01_00.tka
dig.mu2e.FlateminusMix1BBSignal.MDC2020r_best_v1_0.art mcs.mu2e.FlateminusMix1BBSignal.MDC2020z_best_v1_1.art
dig.mu2e.CosmicCRYExtractedCatDigiTrk.MDC2020r.art mcs.mu2e.CosmicCRYExtractedCatDigiTrk.MDC2020z.art nts.mu2e.CosmicCRYExtractedTrk.MDC2020z1_best_v1_1_std_v04_01_00.tka naming issue: best, v1_1

Original datasets; most of these have been supersceded. You must use an appropriate tag of Mu2e code to process these datasets, see the table above

The digitization stream is divided into five parts: Signal, Diag, Trk, Calo, Untriggered, which are described in DocDB 41757. To obtain details (number of events, number of files, GB, etc.) about a dataset you can use the datasetSummary.sh script, available in the Production repository.

In the following summary, [stream] can be Untriggered, Calo, Signal, Diag, Trk and [purpose] can be perfect, best, startup. The digitization [digi] can be Mix[intensity] for mixed samples, OnSpill for unmixed samples. An [intensity] of 1BB corresponds to 1 booster proton batch, 2BB means 2 booster proton batches. MDC2020 also supports 'Low' intensity, but no production samples of that have been produced yet. 'Sequential' mixing is a special case where a fixed pattern of individual proton bunch intensities from a detailed beam slow extraction simulation is used instead of the lognormal intensity generator. Reconstructed (mcs) datasets are currently only available for Signal the stream.

NB: due to a configuration bug the CosmicLivetime data product was dropped from the dig and mcs datasets for MDC2020v production and earlier.

Digi Reco
dig.mu2e.CeEndpoint[digi][stream].MDC2020r_[purpose]_v1_0.art mcs.mu2e.CeEndpoint[digi]Signal.MDC2020r_[purpose]_v1_0.art
dig.mu2e.CePlusEndpoint[digi][stream].MDC2020r_[purpose]_v1_0.art mcs.mu2e.CePlusEndpoint[digi]Signal.MDC2020r_[purpose]_v1_0.art
dig.mu2e.FlateMinus[digi][stream].MDC2020r_[purpose]_v1_0.art mcs.mu2e.FlateMinus[digi]Signal.MDC2020r_[purpose]_v1_0.art
dig.mu2e.FlatePlus[digi][stream].MDC2020r_[purpose]_v1_0.art mcs.mu2e.FlatePlus[digi]Signal.MDC2020r_[purpose]_v1_0.art
dig.mu2e.NoPrimary[digi][stream].MDC2020r_[purpose]_v1_0.art mcs.mu2e.NoPrimary[digi]Signal.MDC2020r_[purpose]_v1_0.art
dig.mu2e.PBINormal_33344MixSeq[stream].MDC2020r_[purpose]_v1_0.art
mcs.mu2e.PBINormal_33344MixSeqSignal.MDC2020r_[purpose]_v1_0.art
dig.mu2e.PBIPathological_33344MixSeq[stream].MDC2020r_[purpose]_v1_0.art mcs.mu2e.PBIPathological_33344MixSeqSignal.MDC2020r_[purpose]_v1_0.art
dig.mu2e.IPAMuminusMichel[digi][stream].MDC2020r_[purpose]_v1_0.art mcs.mu2e.IPAMuminusMichel[digi]Signal.MDC2020r_[purpose]_v1_0.art
dig.mu2e.IPAFlateMinus[digi][stream].MDC2020r_[purpose]_v1_0.art mcs.mu2e.IPAFlateMinus[digi]Signal.MDC2020r_[purpose]_v1_0.art
dig.mu2e.NoPrimaryMix2BB[digi][stream].MDC2020v.art ...
dig.mu2e.CosmicCORSIKAExtractedNoFieldTrk[stream].MDC2020r_[purpose]_v1_0.art (~1 hour of cosmic livetime)
dig.mu2e.CosmicCRYExtractedCatDigi*.MDC2020y_[stream]_v1_1.art (~10 hour of cosmic livetime)

Special datasets

In order to assess the impact of a degraded field from coil 11 in the DS a special CeEndpoint campaign was run. In total 100K CeEndpoints were ran with the updated field map which was passed to the Mu2e/Offline code via the Production scripts. This is currently tagged as MDC2020v. Mixed and OnSpill samples were produced assuming 1BB mode of operation. The datasets with degraded DS field are:

Digi Reco
dig.mu2e.CeEndpoint[digi][stream].MDC2020v_[purpose]_v1_0.art mcs.mu2e.CeEndpoint[digi]Signal.MDC2020v_[purpose]_v1_0.art

Argonne CRY datasets

These are CRY datasets produced by Yuri Oksuzian on the Argonne supercomputer

location records files MB events dataset


  T     2237     2237       559            0   dig.mu2e.CosmicCRYhiOnSpillCalo.MDC2020r_perfect_v1_0.art
  T     2237     2237       559            0   dig.mu2e.CosmicCRYhiOnSpillDiag.MDC2020r_perfect_v1_0.art
  T     2237     2237     10494       137638   dig.mu2e.CosmicCRYhiOnSpillSignal.MDC2020r_perfect_v1_0.art
  T     2237     2237      2275        19387   dig.mu2e.CosmicCRYhiOnSpillTrk.MDC2020r_perfect_v1_0.art
  T     2237     2237     14472       891925   dig.mu2e.CosmicCRYhiOnSpillUntriggered.MDC2020r_perfect_v1_0.art
  T      200      200      1099          169   dig.mu2e.CosmicCRYhiSigFiltMix2BBCalo.MDC2020s_perfect_v1_0.art
  T      200      200       145            0   dig.mu2e.CosmicCRYhiSigFiltMix2BBDiag.MDC2020s_perfect_v1_0.art
  T      200      200    370173        87358   dig.mu2e.CosmicCRYhiSigFiltMix2BBSignal.MDC2020s_perfect_v1_0.art
  T      200      200     39523         9099   dig.mu2e.CosmicCRYhiSigFiltMix2BBTrk.MDC2020s_perfect_v1_0.art
  T      200      200    137872        30371   dig.mu2e.CosmicCRYhiSigFiltMix2BBUntriggered.MDC2020s_perfect_v1_0.art
  T     7897     7896      1964            0   dig.mu2e.CosmicCRYloOnSpillCalo.MDC2020r_perfect_v1_0.art
  T     7897     7897      1964            0   dig.mu2e.CosmicCRYloOnSpillDiag.MDC2020r_perfect_v1_0.art
  T     7897     7897     25841       392295   dig.mu2e.CosmicCRYloOnSpillSignal.MDC2020r_perfect_v1_0.art
  T     7897     7896      5609        45496   dig.mu2e.CosmicCRYloOnSpillTrk.MDC2020r_perfect_v1_0.art
  T     7897     7897     51348      4324567   dig.mu2e.CosmicCRYloOnSpillUntriggered.MDC2020r_perfect_v1_0.art
  T      213      213     14993       126613   mcs.mu2e.CosmicCRYhiOnSpillSignal.MDC2020r_perfect_v1_0.art
  T      196      196     29577        82601   mcs.mu2e.CosmicCRYhiSigFiltMix2BBSignal.MDC2020s_perfect_v1_0.art
  T      781      781     43287       379186   mcs.mu2e.CosmicCRYloOnSpillSignal.MDC2020r_perfect_v1_0.art

Flat Gamma Samples

The following samples were made with these changes to the ``PrimaryFilter":

  • MinimumPartMom : 20.0 # MeV/c (nominal = 50)
  • MinimumTrkSteps : 2 # primary must produce at least this many TrkSteps (nominal = 12)
  • MinimumSumCaloStepE : 5.0 # or at least this much calo energy (nominal = 5)

A Flat photon 70-102 MeV/c spectrum was generated from initial stopped muons, detector steps and digi files are available:

  • dts.mu2e.FlatGamma.MDC2020z_sm3.art
  • dig.mu2e.FlatGammaOnSpillConv.MDC2020z_sm3_perfect_v1_0.art

DIOtail

Two DIOtail test samples are available at dts stage. These use a binned spectrum based on Czarnecki et al.


momentum cut release filename files events
p > 95 MeV/c MDC2020ad dts.mu2e.DIOtail.MDC2020ad.art 99 1534077
p > 75 MeV/c MDC2020ad_sm0 dts.mu2e.DIOtail.MDC2020ad_sm0.art 83 30357435

Beam Pileup datasets

The final production beam pileup (detector signals from beam particles) datasets for normal mixing are listed below. Note that normal workflows do not require you to ever directly access these: mixing jobs to use these should be configured using the Production/Scripts/gen_Mix.sh script.

files     MB      events     dataset

 9598 2948814  874688822   dts.mu2e.NeutralsFlashCat.MDC2020p.art
 9700  211960  248809400   dts.mu2e.EleBeamFlashCat.MDC2020p.art
   50   47481   43638457   dts.mu2e.MuStopPileupCat.MDC2020p.art
   20   15659   22262973   dts.mu2e.MuBeamFlashCat.MDC2020p.art

The NeutralsFlashCat and EleBeamFlash datasets are quite large and require a lot of staging time and space

In addition, there are production background datasets for early digitization mixing listed below. Note there are no 'Early' MuStopPileup, as the standard dataset version of those works fine for early (>200ns) digitization mixing.

files     MB      events     dataset

  179    2279     521442   dts.mu2e.EarlyNeutralsFlashCat.MDC2020p.art
   92     425     113819   dts.mu2e.EarlyEleBeamFlashCat.MDC2020p.art
   19     859    1151125   dts.mu2e.EarlyMuBeamFlashCat.MDC2020p.art

Sim Datasets

Datasets from early stages of production are available for testing, including:

Beam

TODO

Stops

Dataset Description
sim.mu2e.MuminusStopsCat.MDC2020p.art Negative muon stops in the stopping target, used in muon daughter primary and pileup production campaigns
sim.mu2e.IPAMuminusStopsCat.MDC2020r.art Negative muon stops in the IPA, used to produce calibration samples of Michel electrons
sim.mu2e.MuplusStopsCat.MDC2020t.art Positive muon stops in the stopping target. There are only a few thousand of these and they are highly asymmetric

Pions

The pion beam campaign is a separate one from the main muon campaign. The naming convention for the various stages follows that of the muon beam campaign. Two pion beam campaigns were carried out: MDC2020r and MDC2020t

Dataset Description
sim.mu2e.PiminusStopsCat.MDC2020t.art Negative pion stops in the stopping target, pions have an infinite lifetime
sim.mu2e.PiplusStopsCat.MDC2020t.art positive pion stops in the stopping target, pions have an infinite lifetime

Older Datasets

1% and 10% test datasets were produced when testing MDC2020, these still show up in samweb listings but have been superseded and should not be used anymore. This includes datasets with the following descriptions: MDC2020k MDC2020km MDC2020n_10h MDC2020n_10pc