17_11_21_dataMC

data : /MinimumBias[*]/Commissioning2021-PromptReco-v1/* [*]=1,2,...9    , GT : 120X_dataRun3_Prompt_v2

data info : Pileup is same in all samples. The minimum bias trigger had too high rate to populate a single data stream to be handled in one piece at Tier0 which has hard limits on the #events/s throughput, so it was split in 10 streams, but they contain exactly the same kind of events in priciple

MC : /RelValMinBias_14TeV/CMSSW_12_0_2_patch1-121X_mcRun3_2021_realistic_forpp900GeV_v1_JIRA137EraRun3-v2/*  , GT : 121X_mcRun3_2021_realistic_forpp900GeV_v1

CMS release : CMSSW_12_0_3_patch1 (according to data)

JSon : https://cms-service-dqmdc.web.cern.ch/CAF/certification/Collisions21/collisions21Special_346235_346512_DcsTrackerPixelJSON.txt

Suggestions : Also I would recommend you filter out empty events by requiring that this trigger bit corresponding to pixel activity is firing in the data [1]
Please be careful that this trigger is not simulated in MC, so filteringon it in MC I am afraid would not select anything.

```
[1]
# HLT Filter
import HLTrigger.HLTfilters.hltHighLevel_cfi

process.HLTFilter =
HLTrigger.HLTfilters.hltHighLevel_cfi.hltHighLevel.clone(
     HLTPaths = ['HLT_*PixelCluster*'],
     throw = False #dont throw except on unknown path names
)
process.hltFilter_step = cms.Path(process.HLTFilter)
```

Marco : If you are to choose only one run, I would advise to use only runs *after* the LHC cogging change, because the tracking is heavily perturbed by the wrong position of the luminous region.
So please avoid runs from before fill 7517 (actually in fact, all runs before 346389 should be avoided, as they contain also the Pixel timing scan).

Our target runs are : 

"346396": [
[
1,
199
]

"346490": [
[
17,
413
]

"346512": [
[
1,
612
]
]

PU information : 
The MC sample has 0 PU, while the data is basically also 0PU . I think they were expecting PU=0.1 , 1 event every 10 BX