Small improvements in Pixel track reconstruction #14247

VinInn · 2016-04-26T10:20:54Z

relevant almost exclusively for HLT.
No regression expected
(in reco notably as little of the modified code run offline)

PR made now to allow @makortel to build his PhaseI development above it

cmsbuild · 2016-04-26T10:21:16Z

A new Pull Request was created by @VinInn (Vincenzo Innocente) for CMSSW_8_1_X.

It involves the following packages:

RecoPixelVertexing/PixelLowPtUtilities
RecoPixelVertexing/PixelTrackFitting
RecoPixelVertexing/PixelTriplets
RecoTracker/CkfPattern
RecoTracker/TkHitPairs
RecoTracker/TkSeedGenerator
RecoTracker/TkSeedingLayers

@cmsbuild, @cvuosalo, @slava77, @davidlange6 can you please review it and eventually sign? Thanks.
@ghellwig, @makortel, @GiacomoSguazzoni, @rovere, @VinInn, @mschrode, @gpetruc, @cerati, @dgulhan this is something you requested to watch as well.
@slava77, @Degano, @smuzaffar you are the release manager for this.

cms-bot commands are list here #13028

VinInn · 2016-04-26T11:51:53Z

@cmsbuild , please test

VinInn · 2016-04-26T11:52:23Z

@mtosi

cmsbuild · 2016-04-26T12:21:18Z

The tests are being triggered in jenkins.
https://cmssdt.cern.ch/jenkins/job/ib-any-integration/12638/console

cmsbuild · 2016-04-26T14:52:36Z

-1

Tested at: f3e6cdf

You can see the results of the tests here:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-14247/12638/summary.html

I found follow errors while testing this PR

Failed tests: RelVals

RelVals:

When I ran the RelVals I found an error in the following worklfows:
1001.0 step3

runTheMatrix-results/1001.0_RunMinBias2011A+RunMinBias2011A+TIER0EXP+ALCAEXP+ALCAHARVD1+ALCAHARVD2+ALCAHARVD3+ALCAHARVD4/step3_RunMinBias2011A+RunMinBias2011A+TIER0EXP+ALCAEXP+ALCAHARVD1+ALCAHARVD2+ALCAHARVD3+ALCAHARVD4.log

VinInn · 2016-04-26T15:44:57Z

10 10 9 7 5 1 1 tests passed, 0 0 0 0 0 0 0 failed

what shall we do, run tests again?

slava77 · 2016-04-26T15:58:43Z

@cmsbuild please test

ret= 35072
maybe the node had a problem

cmsbuild · 2016-04-26T15:59:39Z

The tests are being triggered in jenkins.
https://cmssdt.cern.ch/jenkins/job/ib-any-integration/12647/console

VinInn · 2016-04-28T12:06:25Z

@Dr15Jones This has already discussed . this is a pattern used everywhere in pixel reco.
It is at least consistent.
Please open a more general issue that proper percolation of magnetic field is required in the WHOLE
RecoPixelVertexing package (and proper optimize, we cannot for each hit ask what is the magfield at 0,0,0)

cmsbuild · 2016-04-28T12:27:46Z

Pull request #14247 was updated. @cmsbuild, @cvuosalo, @slava77, @davidlange6 can you please check and sign again.

VinInn · 2016-04-28T12:27:57Z

@cmsbuild , please test

cmsbuild · 2016-04-28T12:28:19Z

The tests are being triggered in jenkins.
https://cmssdt.cern.ch/jenkins/job/ib-any-integration/12689/console

VinInn · 2016-04-28T12:28:43Z

we should add a multi run test
(I am not sure if HLT keeps the same process open at runs boundaries)

Dr15Jones · 2016-04-28T12:54:19Z

RecoPixelVertexing/PixelTrackFitting/src/PixelFitterByHelixProjections.cc

-  }
-
-  if (theFieldWatcher.check(es)) {
+  {  


This change is fine with me. However, the ESWatcher theFieldWatcher was guaranteeing that theField was being updated on IOV changes.

they were utterly slow

Dr15Jones · 2016-04-28T13:01:14Z

unhold

cmsbuild · 2016-04-28T16:20:35Z

-1

Tested at: 56b5e61

You can see the results of the tests here:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-14247/12689/summary.html

I found follow errors while testing this PR

Failed tests: RelVals AddOn

RelVals:

When I ran the RelVals I found an error in the following worklfows:
134.911 step2

runTheMatrix-results/134.911_RunSinglePh2015D+RunSinglePh2015D+HLTDR2_2016+RECODR2_2016reHLT+HARVESTDR2/step2_RunSinglePh2015D+RunSinglePh2015D+HLTDR2_2016+RECODR2_2016reHLT+HARVESTDR2.log

1306.0 step2

runTheMatrix-results/1306.0_SingleMuPt1_UP15+SingleMuPt1_UP15+DIGIUP15+RECOUP15+HARVESTUP15/step2_SingleMuPt1_UP15+SingleMuPt1_UP15+DIGIUP15+RECOUP15+HARVESTUP15.log

135.4 step1

runTheMatrix-results/135.4_ZEE_13+ZEEFS_13+HARVESTUP15FS+MINIAODMCUP15FS/step1_ZEE_13+ZEEFS_13+HARVESTUP15FS+MINIAODMCUP15FS.log

1330.0 step2

runTheMatrix-results/1330.0_ZMM_13+ZMM_13+DIGIUP15+RECOUP15+HARVESTUP15/step2_ZMM_13+ZMM_13+DIGIUP15+RECOUP15+HARVESTUP15.log

10021.0 step2

runTheMatrix-results/10021.0_TenMuE_0_200+TenMuE_0_200_pythia8_2017_GenSimFull+DigiFull_2017+RecoFull_2017+HARVESTFull_2017/step2_TenMuE_0_200+TenMuE_0_200_pythia8_2017_GenSimFull+DigiFull_2017+RecoFull_2017+HARVESTFull_2017.log

25202.0 step2

runTheMatrix-results/25202.0_TTbar_13+TTbar_13+DIGIUP15_PU25+RECOUP15_PU25+HARVESTUP15_PU25/step2_TTbar_13+TTbar_13+DIGIUP15_PU25+RECOUP15_PU25+HARVESTUP15_PU25.log

10024.0 step2

runTheMatrix-results/10024.0_TTbar_13+TTbar_13TeV_TuneCUETP8M1_2017_GenSimFull+DigiFull_2017+RecoFull_2017+HARVESTFull_2017/step2_TTbar_13+TTbar_13TeV_TuneCUETP8M1_2017_GenSimFull+DigiFull_2017+RecoFull_2017+HARVESTFull_2017.log

AddOn:

I found errors in the following addon tests:

cmsDriver.py RelVal -s L1REPACK:Full --data --scenario=pp -n 10 --conditions auto:run2_hlt_PRef --relval 9000,50 --datatier "RAW" --customise=HLTrigger/Configuration/CustomConfigs.L1T --era Run2_2016 --magField 38T_PostLS1 --eventcontent RAW --fileout file:RelVal_Raw_PRef_DATA.root --filein /store/data/Run2015D/MuonEG/RAW/v1/000/256/677/00000/80950A90-745D-E511-92FD-02163E011C5D.root : FAILED - time: date Thu Apr 28 18:04:58 2016-date Thu Apr 28 18:00:03 2016 s - exit: 16896
cmsDriver.py RelVal -s L1REPACK:Full --data --scenario=HeavyIons -n 10 --conditions auto:run2_hlt_HIon --relval 9000,50 --datatier "RAW" --eventcontent RAW --customise=HLTrigger/Configuration/CustomConfigs.L1T --era Run2_2016,Run2_HI --magField 38T_PostLS1 --fileout file:RelVal_Raw_HIon_DATA.root --filein /store/hidata/HIRun2015/HIHardProbes/RAW-RECO/HighPtJet-PromptReco-v1/000/263/689/00000/1802CD9A-DDB8-E511-9CF9-02163E0138CA.root : FAILED - time: date Thu Apr 28 18:04:29 2016-date Thu Apr 28 18:00:06 2016 s - exit: 16896
cmsDriver.py TTbar_Tauola_13TeV_cfi -s GEN,SIM,DIGI,L1,DIGI2RAW --mc --scenario=pp -n 10 --conditions auto:run2_mc_GRun --relval 9000,50 --datatier "GEN-SIM-RAW" --eventcontent RAWSIM --customise=HLTrigger/Configuration/CustomConfigs.L1T --era Run2_2016 --magField 38T_PostLS1 --fileout file:RelVal_Raw_GRun_MC.root : FAILED - time: date Thu Apr 28 18:08:43 2016-date Thu Apr 28 18:00:10 2016 s - exit: 16896
cmsDriver.py TTbar_Tauola_13TeV_cfi -s GEN,SIM,DIGI,L1,DIGI2RAW --mc --scenario=pp -n 10 --conditions auto:run2_mc_PRef --relval 9000,50 --datatier "GEN-SIM-RAW" --eventcontent RAWSIM --customise=HLTrigger/Configuration/CustomConfigs.L1T --era Run2_2016 --magField 38T_PostLS1 --fileout file:RelVal_Raw_PRef_MC.root : FAILED - time: date Thu Apr 28 18:06:50 2016-date Thu Apr 28 18:00:17 2016 s - exit: 16896
cmsDriver.py TTbar_13TeV_TuneCUETP8M1_cfi --conditions auto:run2_mc --fast -n 100 --eventcontent AODSIM,DQM --relval 100000,1000 -s GEN,SIM,RECOBEFMIX,DIGI:pdigi_valid,L1,DIGI2RAW,L1Reco,RECO,EI,HLT:@relval2016,VALIDATION --customise=HLTrigger/Configuration/CustomConfigs.L1THLT --datatier GEN-SIM-DIGI-RECO,DQMIO --beamspot NominalCollision2015 --era Run2_2016 --magField 38T_PostLS1 : FAILED - time: date Thu Apr 28 18:09:06 2016-date Thu Apr 28 18:00:21 2016 s - exit: 16896
cmsDriver.py RelVal -s L1REPACK:Full --data --scenario=pp -n 10 --conditions auto:run2_hlt_PIon --relval 9000,50 --datatier "RAW" --customise=HLTrigger/Configuration/CustomConfigs.L1T --era Run2_2016 --magField 38T_PostLS1 --eventcontent RAW --fileout file:RelVal_Raw_PIon_DATA.root --filein /store/data/Run2015D/MuonEG/RAW/v1/000/256/677/00000/80950A90-745D-E511-92FD-02163E011C5D.root : FAILED - time: date Thu Apr 28 18:05:11 2016-date Thu Apr 28 18:01:13 2016 s - exit: 16896
cmsDriver.py TTbar_Tauola_13TeV_cfi -s GEN,SIM,DIGI,L1,DIGI2RAW --mc --scenario=HeavyIons -n 10 --conditions auto:run2_mc_HIon --relval 9000,50 --datatier "GEN-SIM-RAW" --eventcontent RAWSIM --customise=HLTrigger/Configuration/CustomConfigs.L1T --era Run2_2016,Run2_HI --magField 38T_PostLS1 --fileout file:RelVal_Raw_HIon_MC.root : FAILED - time: date Thu Apr 28 18:11:24 2016-date Thu Apr 28 18:05:07 2016 s - exit: 16896
cmsDriver.py RelVal -s L1REPACK:Full --data --scenario=pp -n 10 --conditions auto:run2_hlt_GRun --relval 9000,50 --datatier "RAW" --eventcontent RAW --customise=HLTrigger/Configuration/CustomConfigs.L1T --era Run2_2016 --magField 38T_PostLS1 --fileout file:RelVal_Raw_GRun_DATA.root --filein /store/data/Run2015D/MuonEG/RAW/v1/000/256/677/00000/80950A90-745D-E511-92FD-02163E011C5D.root : FAILED - time: date Thu Apr 28 18:14:01 2016-date Thu Apr 28 18:07:40 2016 s - exit: 16896
cmsDriver.py TTbar_Tauola_13TeV_cfi -s GEN,SIM,DIGI,L1,DIGI2RAW --mc --scenario=pp -n 10 --conditions auto:run2_mc_PIon --relval 9000,50 --datatier "GEN-SIM-RAW" --eventcontent RAWSIM --customise=HLTrigger/Configuration/CustomConfigs.L1T --era Run2_2016 --magField 38T_PostLS1 --fileout file:RelVal_Raw_PIon_MC.root : FAILED - time: date Thu Apr 28 18:15:02 2016-date Thu Apr 28 18:08:47 2016 s - exit: 16896

slava77 · 2016-04-28T17:11:13Z

+1

for #14247 56b5e61

the latest update removes caching of the magnetic field pointer, which is logically more stable choice
jenkins test failed due to problems in IB
local tests show no change in performance, based on wf 25202

cmsbuild · 2016-04-28T17:11:37Z

This pull request is fully signed and it will be integrated in one of the next CMSSW_8_1_X IBs (but tests are reportedly failing). This pull request requires discussion in the ORP meeting before it's merged. @slava77, @davidlange6, @Degano, @smuzaffar

Backport of #14247: Small improvements in Pixel track reconstruction

VinInn added 11 commits April 24, 2016 12:08

remove qul check as sone in addResults

edf96e9

reserve, use barPhi

95e1b35

simplify (a bit) PixelTrackReco

4a0095a

speed up pixeltrackcleaner for HLT:|

b64cad1

fix loop

3d28d8c

avoid trigonometric and inversion

94053a0

avoid rechecking lower hits

2e9cbff

set algo, not faster

7726053

ad-hoc set algo

a2af714

at the end this version is the fastest

eb46fe0

compiles

f3e6cdf

cmsbuild added this to the Next CMSSW_8_1_X milestone Apr 26, 2016

cmsbuild added reconstruction-pending pending-signatures tests-pending orp-pending comparison-pending labels Apr 26, 2016

cmsbuild added tests-started and removed tests-pending labels Apr 26, 2016

cmsbuild added tests-rejected and removed tests-started labels Apr 26, 2016

cmsbuild removed the tests-rejected label Apr 26, 2016

in case multiple runs are processed in the same job

56b5e61

cmsbuild added reconstruction-pending tests-pending comparison-pending and removed comparison-available reconstruction-approved tests-approved labels Apr 28, 2016

cmsbuild added tests-started and removed tests-pending labels Apr 28, 2016

Dr15Jones reviewed Apr 28, 2016
View reviewed changes

cmsbuild removed the hold label Apr 28, 2016

cmsbuild added tests-rejected and removed tests-started labels Apr 28, 2016

cmsbuild added reconstruction-approved fully-signed and removed pending-signatures reconstruction-pending labels Apr 28, 2016

davidlange6 merged commit 20d2f64 into cms-sw:CMSSW_8_1_X Apr 29, 2016

fwyzard mentioned this pull request Jul 8, 2016

backport PRs to CMSSW 8.0.x to improve HLT performance #15151

Closed

9 tasks

makortel mentioned this pull request Jul 26, 2016

Backport of #14247: Small improvements in Pixel track reconstruction #15288

Merged

davidlange6 added a commit that referenced this pull request Aug 26, 2016

Merge pull request #15288 from makortel/backport80_pr14247

28387be

Backport of #14247: Small improvements in Pixel track reconstruction

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Small improvements in Pixel track reconstruction #14247

Small improvements in Pixel track reconstruction #14247

VinInn commented Apr 26, 2016

cmsbuild commented Apr 26, 2016

VinInn commented Apr 26, 2016

VinInn commented Apr 26, 2016

cmsbuild commented Apr 26, 2016 •

edited

cmsbuild commented Apr 26, 2016

VinInn commented Apr 26, 2016

slava77 commented Apr 26, 2016

cmsbuild commented Apr 26, 2016 •

edited

VinInn commented Apr 28, 2016

cmsbuild commented Apr 28, 2016

VinInn commented Apr 28, 2016

cmsbuild commented Apr 28, 2016 •

edited

VinInn commented Apr 28, 2016

Dr15Jones Apr 28, 2016

VinInn Apr 28, 2016

Dr15Jones commented Apr 28, 2016

cmsbuild commented Apr 28, 2016

slava77 commented Apr 28, 2016

cmsbuild commented Apr 28, 2016

Small improvements in Pixel track reconstruction #14247

Small improvements in Pixel track reconstruction #14247

Conversation

VinInn commented Apr 26, 2016

cmsbuild commented Apr 26, 2016

VinInn commented Apr 26, 2016

VinInn commented Apr 26, 2016

cmsbuild commented Apr 26, 2016 • edited

cmsbuild commented Apr 26, 2016

VinInn commented Apr 26, 2016

slava77 commented Apr 26, 2016

cmsbuild commented Apr 26, 2016 • edited

VinInn commented Apr 28, 2016

cmsbuild commented Apr 28, 2016

VinInn commented Apr 28, 2016

cmsbuild commented Apr 28, 2016 • edited

VinInn commented Apr 28, 2016

Dr15Jones Apr 28, 2016

Choose a reason for hiding this comment

VinInn Apr 28, 2016

Choose a reason for hiding this comment

Dr15Jones commented Apr 28, 2016

cmsbuild commented Apr 28, 2016

slava77 commented Apr 28, 2016

cmsbuild commented Apr 28, 2016

cmsbuild commented Apr 26, 2016 •

edited

cmsbuild commented Apr 26, 2016 •

edited

cmsbuild commented Apr 28, 2016 •

edited