Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[106X] Remove JetBTagInfo class #1544

Merged
merged 3 commits into from
Feb 12, 2021
Merged

[106X] Remove JetBTagInfo class #1544

merged 3 commits into from
Feb 12, 2021

Conversation

raggleton
Copy link
Contributor

Removed since it has a lot of unused memory assigned to it in each Jet collection (see e.g. https://uhh2-integration.web.cern.ch/UHH2integration/test1535/mc_UL18_TTSemiLeptonic.html#sizejetsAk4Puppi) and was only previously used for subjet b-tagging SFs/calibrations (I think?)

Removed the header, and references to it in classes.h/classes_def.xml, to tidy up.

This should save a fair amount of space (5.9 -> 5MB on 500 TTBar events using ntuplewriter_mc_UL18.py, let's see how it does on other samples)

Robin Aggleton added 2 commits January 14, 2021 15:49
Remove JetBTagInfo member from Jet class
48f4336 
Also removes it from being potentially filled in NTupleWriter
(It was turned off anyway)

This should save a fair amount of space
(5.9 -> 5MB on 500 TTBar events using ntuplewriter_mc_UL18.py)
Remove JetBTagInfo header, and from classes.h/classes_def.xml
b175839
@uhh2-bot
Copy link
Collaborator

Starting setup & tests, see https://gitlab.cern.ch/raggleto/UHH2-integration/-/pipelines/2231875

@uhh2-bot uhh2-bot added TestsRunning Automated compilation & tests running RunII_106X_v1 labels Jan 14, 2021
@uhh2-bot
Copy link
Collaborator

Compilation with PR successful

@raggleton raggleton added the PleaseTest Manually trigger automated compiling & testing label Jan 14, 2021
@uhh2-bot uhh2-bot removed the PleaseTest Manually trigger automated compiling & testing label Jan 14, 2021
@raggleton raggleton added the PleaseTest Manually trigger automated compiling & testing label Jan 17, 2021
@uhh2-bot uhh2-bot added Passed Automated compilation & testing successful and removed PleaseTest Manually trigger automated compiling & testing TestsRunning Automated compilation & tests running labels Jan 17, 2021
@raggleton raggleton added the PleaseTest Manually trigger automated compiling & testing label Jan 18, 2021
@uhh2-bot
Copy link
Collaborator

Starting setup & tests, see https://gitlab.cern.ch/raggleto/UHH2-integration/-/pipelines/2239128

@uhh2-bot uhh2-bot added TestsRunning Automated compilation & tests running and removed PleaseTest Manually trigger automated compiling & testing Passed Automated compilation & testing successful labels Jan 18, 2021
@uhh2-bot
Copy link
Collaborator

Compilation with PR successful

@uhh2-bot uhh2-bot removed the TestsRunning Automated compilation & tests running label Jan 18, 2021
@uhh2-bot
Copy link
Collaborator

uhh2-bot commented Jan 18, 2021 •
edited by raggleton

Report for PR 1544
Webpages with full plots, timing & size info: https://uhh2-integration-results.web.cern.ch/uhh2-integration-results/UHH2integration/test1544/

Test samples defined here: https://gitlab.cern.ch/raggleto/UHH2-integration/blob/test1544/scripts/cmsrun_jobs.py

Ntuple comparison report

name total # hists added collections added hists removed collections removed hists diff entries no entries zero rms zero value same
data_UL16postVFP_JetHT 2572 0 0 0 574 0 188 496 174 1140
data_UL16postVFP_SingleElectron 2572 0 0 0 574 0 496 440 153 909
data_UL16postVFP_SingleMu 2572 0 0 0 574 0 510 441 151 896
data_UL16preVFP_JetHT 2572 0 0 0 574 0 314 470 163 1051
data_UL16preVFP_SingleElectron 2572 0 0 0 574 0 300 471 167 1060
data_UL16preVFP_SingleMu 2572 0 0 0 574 0 328 479 165 1026
data_UL17_JetHT 2572 0 0 0 574 0 76 517 181 1224
data_UL17_SingleElectron 2572 0 0 0 574 0 510 423 154 911
data_UL17_SingleMu 2572 0 0 0 574 0 538 422 150 888
data_UL18_EGamma 2572 0 0 0 574 0 440 459 159 940
data_UL18_JetHT 2572 0 0 0 574 0 62 530 187 1219
data_UL18_SingleMu 2572 0 0 0 574 0 482 463 156 897
mc_UL16postVFP_TTbar 2368 0 0 0 574 0 87 451 164 1092
mc_UL16preVFP_TTbar 2368 0 0 0 574 0 87 451 163 1093
mc_UL17_TTSemiLeptonic 2368 0 0 0 574 0 87 448 165 1094
mc_UL18_TTSemiLeptonic 2368 0 0 0 574 0 87 449 166 1092
  • diff entries: Differing number of entries
  • no entries: Both hists has 0 entries
  • zero rms: One or both RMSs are 0: stores same value
  • zero value: One or both hists have only 0s
  • same: Histograms are the same (lowest priority)

Timing report

Sample Reference event loop real/event [s] PR event loop real/event [s] diff
data_UL16postVFP_JetHT 0.830 0.635 -0.196 / -23.56 %
data_UL16postVFP_SingleElectron 0.430 0.405 -0.025 / -5.85 %
data_UL16postVFP_SingleMu 0.435 0.417 -0.018 / -4.05 %
data_UL16preVFP_JetHT 0.530 0.498 -0.032 / -6.08 %
data_UL16preVFP_SingleElectron 0.457 0.416 -0.041 / -9.00 %
data_UL16preVFP_SingleMu 0.367 0.348 -0.019 / -5.18 %
data_UL17_JetHT 0.609 0.679 0.070 / 11.46 %
data_UL17_SingleElectron 0.442 0.451 0.008 / 1.91 %
data_UL17_SingleMu 0.434 0.443 0.008 / 1.95 %
data_UL18_EGamma 0.535 0.562 0.027 / 5.08 %
data_UL18_JetHT 0.804 0.815 0.011 / 1.35 %
data_UL18_SingleMu 0.471 0.483 0.012 / 2.46 %
mc_UL16postVFP_TTbar 0.654 0.627 -0.027 / -4.08 %
mc_UL16preVFP_TTbar 0.580 0.589 0.009 / 1.64 %
mc_UL17_TTSemiLeptonic 0.691 0.639 -0.052 / -7.52 %
mc_UL18_TTSemiLeptonic 0.696 0.618 -0.078 / -11.24 %

Size report
(kB = kilobytes, period is a decimal point not thousands separator)

Sample Reference total branch size per event [kB] PR total branch size per event [kB] diff
data_UL16postVFP_JetHT 10.235 6.880 -3.355 / -32.78 %
data_UL16postVFP_SingleElectron 5.457 4.461 -0.997 / -18.26 %
data_UL16postVFP_SingleMu 5.532 4.578 -0.954 / -17.24 %
data_UL16preVFP_JetHT 6.803 5.225 -1.579 / -23.20 %
data_UL16preVFP_SingleElectron 5.940 4.790 -1.151 / -19.37 %
data_UL16preVFP_SingleMu 4.706 3.812 -0.894 / -19.00 %
data_UL17_JetHT 9.787 6.940 -2.847 / -29.09 %
data_UL17_SingleElectron 5.828 4.892 -0.936 / -16.07 %
data_UL17_SingleMu 5.867 4.932 -0.936 / -15.95 %
data_UL18_EGamma 7.529 6.258 -1.271 / -16.88 %
data_UL18_JetHT 11.027 8.834 -2.193 / -19.88 %
data_UL18_SingleMu 6.464 5.392 -1.072 / -16.58 %
mc_UL16postVFP_TTbar 9.190 7.543 -1.647 / -17.92 %
mc_UL16preVFP_TTbar 9.271 7.553 -1.718 / -18.53 %
mc_UL17_TTSemiLeptonic 11.812 10.141 -1.670 / -14.14 %
mc_UL18_TTSemiLeptonic 11.786 10.078 -1.709 / -14.50 %

@uhh2-bot uhh2-bot added the Passed Automated compilation & testing successful label Jan 18, 2021
@raggleton
Copy link
Contributor Author

Looks like removing the JetBTagInfo attribute can save a fair amount of space, ~15 - 20%. Unless anyone actually needs this info, I'd be strongly in favour of removing it.

(The change in timings is uncorrelated with this change, and is just the natural variation across worker nodes)

@mschrode
Copy link
Contributor

The object/info was used in the past for btag calibration purposes, but it has not been used since a while. Conclusion from the user side is that it can be removed.

@mschrode mschrode merged commit 3655869 into UHH2:RunII_106X_v1 Feb 12, 2021
@mschrode mschrode mentioned this pull request Apr 23, 2021
Open
35 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
Passed Automated compilation & testing successful RunII_106X_v1
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
3 participants
@raggleton @uhh2-bot @mschrode