sPHENIX MAPS tracker discussion

[Hideki Okawa <Hideki.Okawa AT cern.ch>]

Heavy Flavor Topical Group Meeting (2021/2/8)

https://indico.bnl.gov/event/10775/

Introduction - Jin (BNL)

Please post on the MDC-1 MatterMost channel for updates

and questions:

https://chat.sdcc.bnl.gov/sphenix/channels/hf-mdc1 

Sebastian (ISU) joined sPHENIX last week and will join

the HF activities. He has started looking into D*. He may

also contribute to jet studies.

MDC1 Status & Plans - Cameron (LANL)

Job submission restarted twice for the ACTS improvement

and vertex map association bugfix. The number of events

per job was reduced from 2k to 1k due to the condor time

limit. 

Jin: Nice to see millions of events showing up every day.

      HF samples will be the first ones to appear.

Hideki: Will the HF samples reach the pT region that would

      create some low-pT jets? It would be useful to test the

      jet framework at least.

Jin: The bb samples should reach pT~10 GeV.

Yasser: What is the pThard range?

Jin: Set to minimum for pThard and no upper limit. You can

      have a look at the config file here:

      https://github.com/sPHENIX-Collaboration/MDC1/blob/main/submit/HF_pp200_signal/rundir/phpythia8_HF_MDC1.cfg

Ming: What is the matching efficiency to TPC tracks?

Jin: When requiring 2 or more MVTX hits, the matching eff.

      is ~90%. It is really good.

Yasser: How many D0's will we get?

Cameron: The samples are filtered for ccbar, but I don't

     have the explicit numbers of D0.

Discussions for HF Track Trigger  - Cameron, Ming, Yasser (LANL)

The strategy is to consider high-pT tracks, track displacement

and event multiplicity.

Jin: Vertex reconstruction is difficult to do in real time.

      However, the collision points are stable at RHIC.

Ming, Jin: 3D DCA is challenging at the hardware level. 2D

      is possible.  Information from TPC is slow. It records

      130 beam crossings per integration time -> ~1 ccbar

      event. The TPC window translates to a minimum rejection.

Jin: There is another strategy to read out everything and

      implement HF track triggers as HLT using CPU or GPU.

Ming: That is an attractive option. One limitation is that

      there will be no calorimeter information. It would be

      great if the calorimeter can do continuous readout.

Sebastian: Are e, mu triggers possible?

Jin: There will be no good mu trigger. There is no muon

      detector and using the HCAL would be difficult. For the

      single e, we can go down to pT~5 GeV but not lower.

b-jet studies  - Dan (UC Boulder)

 

Jin: ACTS was updated a few days ago and should be much

      more stable. Let's use the ACTS version.

Sanghoon: Does ACTS look for multiple or single vertices?

Yasser: It keeps all vertices with 2 or more tracks.

Hideki: Could you upload your macro somewhere so that we

      can follow up?

Dan: Will do on MatterMost.

Sebastian: What kind of jets do we use?

Jin: They are reconstructed by anti-kt.

Hideki: In the MDC-1 samples, jets with various R (from

      0.2 and larger) and inputs (Tower, Topocluster, PFlow)

      are stored.

Jin: anti-kt Tower jets (R=0.4) should be good to start from.

Sebastian: We should check the dependence on the cone size

      and hadron pT.

Hideki: Indeed and we should have sufficient information to

      check that in the MDC-1 samples. Regarding the track-jet

      association, it is not urgent, but the ghost association is

      more precise than the dR matching.

Jin: We should have a dedicated discussion session on jets at

      the next meeting or the one after it.