sPHENIX MAPS tracker discussion

["Huang, Jin" <jhuang AT bnl.gov>]

Introduction - Hideki

 

Discussing joint meeting with Upsilon TG, starting by exploring joint meetings to exchange tools (e.g. KFParticle_sPHENIX) and techniques (e.g. event mixing). Expect aDoodle poll to get a new meeting time. 

 

 

 

MDC1 Vertexing: RAVE vs ACTS - Sanghoon, Woohyeong

Summary: direct cross-check the performance of the MDC1 built-in RAVE and ACTS vertex finders

 

Jin/Joe: expected SvtxVertexMap and SvtxVertexMapACTS to show the same content (current behavior of tracking software). Supprising they look different from MDC1 DST

 

 

Joe: ACTS would be sensitive to tails in the secondary vertex contribution to the primary vertex. 

Sanghoon: reconstruct vertex again with a different filter on ACTS? 

Joe: yes. ACTS vertexing module can take arbitrary vertex maps. So the users could do track filtering to fill a new map and rerun the ACTS vertex finder on it. 

Cameron: Our trick at LHCb for HF was to calculate the difference in Chi2 of the primary vertex with and without that track, if the difference was > 9 that track was tagged as a secondary. 

Cameron: LHCb paper on PV reconstruction for HF

https://cds.cern.ch/record/1756296/files/LHCb-PUB-2014-044.pdf

 

Ming: it may be interesting to compare the distribution of tracks from the secondary vs primary vertex. Worry is that the bb sample may have high multiplicity in the primary vertex

 

Jin: It appears the 2017 jet sample cut on one MVTX hit in vertex finding. It is not used in the RAVE vertex finding in the MDC1, which could count for some of the difference. 

TODO: please test rerun RAVE and ACTS vertexing again with on MVTX hit

 

 

Tracking Status- Joe Osborn

Summary: comprehensive status update on the sPHENIX tracking software stack

 

Wei: do ACTS recover efficiency in the track fitting stage?

ACTS: track fitter of the current ACTS module only performs the fit. 

 

Ming: does the beam axis used in silicon seeding

Joe: not yet used as the beam transverse profile is rather wide (200um).

Ming: why many mismatches if MVTX propagate to INTT

Joe: MVTX has a short level arm and INTT strips have long z-length

 

Cameron: KFparticle inputs are primary vertex, primary and secondary tracks. 

Joe: plan to take far silicon seed from beamline aside as secondary candidates -> rest tracklets to find primary track and vertex -> find secondary track

Jin: To salvage the MDC1 file, we might consider rerun ACTS track finding with truth secondary vertex

Cameron: we could handle tracks in secondary/primary maps. e.g. prompt J/psi you would want a primary track map, D0 you would only want the secondaries track map, and D* you would want to combine both

 

Ming/Wei: beam spot could be useful to reject obvious displace tracks from charm/bottom decays

 

Jin: should we consider three track fits for each track with K, pi, proton candidate assumptions?

Joe: probably OK. Memory should be OK. ACTS fit should be fast too. 

Jin: we need to explicitly demonstrate the need, and we might have only the low pT region that needs this treatment. 

 

Joe: does KFParticle take a full field map?

Cameron: not yet. 

 

Jin: Shall we use the RAVE vertex in the current analysis? 

Joe: Could run RAVE for analysis. Please also consider run the ACTS vertexing on user-defined QA cuts on tracks. 

 

Jin: production start for MB sample 

Joe: 15% pT resolution degraded related to recent changes in the ACTS handling of measurements. 

Cameron: This problem was not there during the MDC1 charm/bottom sample generation. 

Jin/Cameron: take the hit and start the MB production soon. 

 

TODO: check whether field consistency resolve momentum scale offset

TODO: if the above check passes, start production for the MB sample

 

 

 

 

 

 

______________________________

 

Jin HUANG

 

Physicist, Ph.D.

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________

 

From: Huang, Jin
Sent: Sunday, June 27, 2021 10:32 PM
To: sphenix-hf-jets-l AT lists.bnl.gov
Cc: sphenix-physics-l AT lists.bnl.gov; sphenix-maps-l AT lists.bnl.gov
Subject: [June 28 Monday 8PM EDT] bi-weekly heavy flavor topical group meeting

 

Hi Everyone,

 

This is a reminder that we will have our next HF topical group meeting on June 28 Monday 8PM EDT.  

 

In this meeting, we have invited the Joe Osborn from the tracking team to join us to talk about the status of tracking software and its implications to the MDC1 HF analysis, preparations for the coming productions, and Q&As.

 

Please join us via

 

• Indico: https://indico.bnl.gov/event/12275/
• ZOOM connection: https://bnl.zoomgov.com/j/1613938007?pwd=QjYzUnpYQ2xNZWZLSy9qd3Z4dFpzZz09

 

We look forward to talking to you all at the meeting.

 

Best regards,

 

Hideki & Jin

 

 

 

______________________________

 

Jin HUANG

 

Physicist, Ph.D.

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________