PFlow reconstruction in Heavy-Ions conditions

This MR has the purpose of providing the right inputs to PFlow algorithm under Heavy-Ion reconstruction conditions

In HeavyIon mode of reco, we name the calorimeter objects differently once we apply the energy subtraction, e.g. "AllCalo" --> "SubtractedCells", "CaloCalTopoClusters" --> "SubtractedCaloCalTopoClusters".. and so on

So far I have seen two places where calorimeter objects are read in the PFlow context:

• PFCfg.py#0350, here cells key is provided through the "Egamma.Keys" then is updated automatically when Heavy-Ion egamma is ON, no change needed.
• PFCfg.py#0420, here "CaloTopoClusters" and "CaloCalTopoClusters" are read. --> Adding functionality to read SubtractedCaloTopoCluster when HImode reco is ON

@mhodgkin would like to comment on this?

Is there any other place you think we should check that the inputs are right?

Also tagging HI people interested in PFlow and conveners, @mrybar @qhu @dhangal @angerami For reconstruction @jdandoy @christos

added 23.0 HeavyIon JetEtmiss Reconstruction labels

Some more information.

By making this change, I see a reduction in the "JetETMissNeutralParticleFlowObjects" container size which makes sense. (JetETMissChargedParticleFlowObjects do not change)

you can clearly see the change in the number of NeutralPFlow objects and the reduction of energy in the figure. RED using Standard TopoClusters, GREEN using SubtractedTopoClusters.

I think its ok...but the standard inputs don't exist in HI so it should not work at all if you don't migrate everything?

Cheers,

Mark

would suggest you grep ReadHandle in all the eflowRec header files too...

Just a quick comment on the overall strategy-- I doubt the physics performance will be great here as the E/p will need to be recalculated for the conditions. Is there a plan for that?

resolved all threads

Probably also true of e.g cuts that turn pflow off in dense environments.

Naively I would expect the algorithm needs retuning for the HI environment to give the best performance.

@dhangal, @qhu or @angerami can you comment on the plan & performance with pflow objects?

We performed extensive studies on topoclusters and TCCs in heavy-ion collisions, the summary of which are detailed in this HCW2022 presentation by @qhu https://indico.cern.ch/event/1167019/sessions/450264/attachments/2500708/4296328/HCW_HIJetConst_QHu_1_9_2022.pdf

As @mswiatlo and @mhodgkin mentioned, the performance of PFlow objects is not expected to be optimal out of the box for heavy-ion collisions without any tuning. However, we would still like to study the behavior and kinematics of these objects in HIC to understand what the tuning should look like.

resolved all threads

marked this merge request as ready

Jenkins please retry a build

This merge request affects 1 package:

• Reconstruction/eflowRec

This merge request affects 1 file:

• Reconstruction/eflowRec/python/PFCfg.py

Adding @goetz ,@mhodgkin ,@cantel as watchers

added review-pending-level-1 label

CI Result SUCCESS (hash da0d08b6)

	Athena
externals
cmake
make
tests

Full details available on this CI monitor view. Check the JIRA CI status board for known problems
Athena: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-CC7 78046]

The changes are straightforward and there are no issues with the CI. Approving from L1.

added review-approved label and removed review-pending-level-1 label

added review-approved-tier0 label

@stapiaar - is this offline reco only or can affect trigger?

resolved all threads

resolved all threads

mentioned in commit 614546f3

merged

mentioned in merge request !65910 (merged)