sPHENIX tracking discussion

[Anthony Frawley <afrawley AT fsu.edu>]

Hi Chris,

To make sure I understand correctly:

For 80-90 event jobs, the time is ~ 1100 s => 13 s / event.

The (I assume peak from the log files?) memory is 7.7 GB.

Right?

Thanks

Tony

---

From: sPHENIX-tracking-l <sphenix-tracking-l-bounces AT lists.bnl.gov> on behalf of pinkenburg via sPHENIX-tracking-l <sphenix-tracking-l AT lists.bnl.gov>
Sent: Wednesday, December 15, 2021 1:18 PM
To: sphenix-tracking <sphenix-tracking-l AT lists.bnl.gov>
Subject: [Sphenix-tracking-l] tracking tests

 

Hi folks,

I am playing with our new hardware using the current new build - it
works pretty nice, for our last step which resembles the "real" tracking
it goes through 10k MB minbias hijing events within a few hours. The
memory and time per job is attached (it is about 80-90 events per file,
we drop foreground events from outside our vertex cut), and... no more
double/tripple peaks.

I do see two type of (rare) crashes which are reproducible. We don't
have random number generators at that point and you can just run our
Fun4All_G4_Trkr.C macro (just copy it to your macro/detector/sPHENIX
checkout - it uses the generic G4Setup_sPHENIX.C in there):
https://urldefense.com/v3/__https://github.com/sPHENIX-Collaboration/MDC2/blob/main/submit/fm_0_20/pass4trk/rundir/Fun4All_G4_Trkr.C__;!!PhOWcWs!iiLYaWAxCh5xT4J77cLvDyMyejoiCNekG085deX7LRMNuvgEdg-646jclqJGwOWq$

Just run it like:
root.exe -q -b
Fun4All_G4_Trkr.C\(0,\"DST_TRKR_HIT_sHijing_0_20fm_50kHz_bkg_0_20fm-0000000002-08417.root\"\)



DST_TRKR_HIT_sHijing_0_20fm_50kHz_bkg_0_20fm-0000000002-03116.root
results in
root.exe:
/cvmfs/sphenix.sdcc.bnl.gov/gcc-8.3/release/release_new/new.2/include/Acts/EventData/SingleBoundTrackParameters.hpp:61:
Acts::SingleBoundTrackParameters<charge_t>::SingleBoundTrackParameters(std::shared_ptr<const
Acts::Surface>, const ParametersVector&,
Acts::SingleBoundTrackParameters<charge_t>::Scalar,
std::optional<Eigen::Matrix<double, 6, 6, 0, 6, 6> >) [with charge_t =
Acts::SinglyCharged;
Acts::SingleBoundTrackParameters<charge_t>::ParametersVector =
Eigen::Matrix<double, 6, 1, 0, 6, 1>;
Acts::SingleBoundTrackParameters<charge_t>::Scalar = double]: Assertion
`(0 <= (params[eBoundQOverP] * q)) and "Inconsistent q/p and q signs"'
failed.

DST_TRKR_HIT_sHijing_0_20fm_50kHz_bkg_0_20fm-0000000002-03170.root
DST_TRKR_HIT_sHijing_0_20fm_50kHz_bkg_0_20fm-0000000002-04880.root
DST_TRKR_HIT_sHijing_0_20fm_50kHz_bkg_0_20fm-0000000002-05015.root
DST_TRKR_HIT_sHijing_0_20fm_50kHz_bkg_0_20fm-0000000002-07310.root
DST_TRKR_HIT_sHijing_0_20fm_50kHz_bkg_0_20fm-0000000002-08417.root
result in
/home/phnxbld/sPHENIX/gcc-8.3/new/source/coresoftware/offline/framework/fun4all/Fun4AllServer.cc:586:
caught exception thrown during process_event from PHActsVertexPropagator
error: No fitted track parameters for trajectory with entry index = 23
where the index is naturally  different.

Basically running tracking over 1M events whenever you want to try
something is no problem.

Chris

--
*************************************************************

Christopher H. Pinkenburg       ;    pinkenburg AT bnl.gov
                                ;    https://urldefense.com/v3/__http://www.phenix.bnl.gov/*pinkenbu__;fg!!PhOWcWs!iiLYaWAxCh5xT4J77cLvDyMyejoiCNekG085deX7LRMNuvgEdg-646jclt_lmA5x$

Brookhaven National Laboratory  ;    phone: (631) 344-5692
Physics Department Bldg 510 C   ;    fax:   (631) 344-3253
Upton, NY 11973-5000

*************************************************************