sphenix-software-l AT lists.bnl.gov
Subject: sPHENIX discussion of software
List archive
[Sphenix-software-l] HI jet reco software & performance
- From: "Perepelitsa, Dennis" <dvp AT bnl.gov>
- To: "sphenix-jet-structure-l AT lists.bnl.gov" <sphenix-jet-structure-l AT lists.bnl.gov>, "sphenix-software-l AT lists.bnl.gov" <sphenix-software-l AT lists.bnl.gov>
- Subject: [Sphenix-software-l] HI jet reco software & performance
- Date: Mon, 12 Jun 2017 01:05:41 +0000
Hello heavy ion jet enthusiasts,
As discussed in our last Jet Structure Topical Group meeting (and as you may have seen in subsequent Simulations meetings and automated git notifications), we have now mostly re-implemented the “ATLAS-style” iterative background determination
and subtraction capabilities for HI jet reconstruction, which did not survive the PHENIX/sPHENIX software “divorce”. I say “mostly" because we are hoping to implement two last pieces of the full functionality (an additional intermediate iteration step to refine
the selection of jets used to define exclusion regions, and the flow estimation / modulation) soon.
The procedure follows the method laid out in the PRC methods paper: https://arxiv.org/abs/1203.1353 (with the two to-be-implemented exceptions above). It determines the mean UE contribution
in each of the three calorimeter subsystems separately in eta=0.1-sized strips while excluding regions which are consistent with high-pT jets from that estimation (in this case, the exclusion regions are set by the positions of R=0.2 pT > 25 GeV jets run without
subtraction). It then creates UE-subtracted versions of calorimeter towers in those subsystems and runs jet reconstruction on this input — taking care to set negative-E towers to have a small positive energy for the purposes of clustering and then update the
final jet kinematics afterwards.
The core code lives in offline/packages/jetbackground: https://github.com/sPHENIX-Collaboration/coresoftware/tree/master/offline/packages/jetbackground
And you can find an example of how to configure and run the HI jet reco here: https://github.com/dvperepelitsa/macros/blob/CaloTriggerDev/macros/g4simulations/G4_HIJetReco.C (which
I also expect will become part of the default Fun4All simulations macro soon)
As a first test of this functionality, Erin Bossard, a student working with us at CU-Boulder, has studied the jet performance systematically in R size and jet pT in p+p and Hijing Au+Au b=0-4fm collisions. I have collected some material on our
subgroup page on the sPHENIX Wiki here: https://wiki.bnl.gov/sPHENIX/index.php/Jet_Structure_Topical_Group , and I attach some of the key plots in a simple set of slides
to this email.
Please do not hesitate with any questions or suggestions — of course, Rosi will also discuss these developments during our report on Tuesday in the Collaboration Meeting.
Dennis
Dennis V. Perepelitsa
Assistant Professor, Physics Department
University of Colorado Boulder
Attachment:
dvp-sPHENIX-JetPerformance-6-11-17.pdf
Description: dvp-sPHENIX-JetPerformance-6-11-17.pdf
- [Sphenix-software-l] HI jet reco software & performance, Perepelitsa, Dennis, 06/11/2017
Archive powered by MHonArc 2.6.24.