Improve the memory usage in the alpaka pixel reconstruction #44458

fwyzard · 2024-03-18T22:09:23Z

PR description:

Fix the TrackingRecHitsDevice constructor:
Fix the hitsModuleStart copy to use the correct source (device-to-device instead of host-to-device).
Improve the offsetBPIX2 copy and use the data member as source, to guarantee the lifetime of the source.
Remove the queue synchronisation.
Remove an unused variable from SiPixelRawToCluster::acquire().
Improve memory usage in SiPixelRawToCluster::acquire():
Allocate device memory only when actually used.
Use cached memory buffers in WordFedAppender.

PR validation:

Run the HLT 2024 v1.0 menu with the changes.

Backport status

To be backported to CMSSW 14.0.x for data taking (see #44459).

fwyzard · 2024-03-18T22:09:30Z

type bugfix

fwyzard · 2024-03-18T22:09:33Z

enable gpu

fwyzard · 2024-03-18T22:09:36Z

please test

cmsbuild · 2024-03-18T22:09:48Z

cms-bot internal usage

cmsbuild · 2024-03-18T22:16:39Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-44458/39532

This PR adds an extra 28KB to repository
There are other open Pull requests which might conflict with changes you have proposed:
- File DataFormats/TrackingRecHitSoA/interface/TrackingRecHitsDevice.h modified in PR(s): Alpaka vs CUDA DQM compare modules for pixel tracks objects #43964

cmsbuild · 2024-03-18T22:17:01Z

A new Pull Request was created by @fwyzard for master.

It involves the following packages:

DataFormats/TrackingRecHitSoA (reconstruction, heterogeneous)
RecoLocalTracker/SiPixelClusterizer (reconstruction)

@jfernan2, @fwyzard, @makortel, @mandrenguyen can you please review it and eventually sign? Thanks.
@mroguljic, @JanFSchulte, @mmusich, @ferencek, @GiacomoSguazzoni, @threus, @felicepantaleo, @rovere, @VinInn, @gpetruc, @VourMa, @dkotlins, @mtosi, @tsusa, @missirol, @tvami this is something you requested to watch as well.
@rappoccio, @sextonkennedy, @antoniovilela you are the release manager for this.

cms-bot commands are listed here

Fix the hitsModuleStart copy to use the correct source (device-to-device instead of host-to-device). Improve the offsetBPIX2 copy and use the data member as source, to guarantee the lifetime of the source. Remove the queue synchronisation. Update the unit test accordingly.

Co-authored-by: Matti Kortelainen <[email protected]>

Allocate device memory only when actually used. Co-authored-by: Matti Kortelainen <[email protected]>

Co-authored-by: Matti Kortelainen <[email protected]>

fwyzard · 2024-03-18T23:27:35Z

please test

fwyzard · 2024-03-20T14:04:53Z

please test

cmsbuild · 2024-03-20T14:08:19Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-44458/39593

This PR adds an extra 36KB to repository
There are other open Pull requests which might conflict with changes you have proposed:
- File DataFormats/TrackingRecHitSoA/interface/TrackingRecHitsDevice.h modified in PR(s): Alpaka vs CUDA DQM compare modules for pixel tracks objects #43964, Update of the Phase-2 GMT Emulator #44319
- File DataFormats/TrackingRecHitSoA/test/alpaka/Hits_test.cc modified in PR(s): Update of the Phase-2 GMT Emulator #44319
- File HeterogeneousCore/AlpakaInterface/interface/OneToManyAssoc.h modified in PR(s): Update of the Phase-2 GMT Emulator #44319
- File HeterogeneousCore/AlpakaInterface/test/alpaka/testPrefixScan.dev.cc modified in PR(s): Update of the Phase-2 GMT Emulator #44319
- File RecoLocalTracker/SiPixelClusterizer/plugins/alpaka/SiPixelRawToClusterKernel.h modified in PR(s): Update of the Phase-2 GMT Emulator #44319

cmsbuild · 2024-03-20T14:08:43Z

Pull request #44458 was updated. @fwyzard, @makortel can you please check and sign again.

cmsbuild · 2024-03-20T17:47:50Z

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-837ba7/38304/summary.html
COMMIT: 70494ae
CMSSW: CMSSW_14_1_X_2024-03-20-1100/el8_amd64_gcc12
Additional Tests: GPU
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/44458/38304/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

You potentially added 88 lines to the logs
Reco comparison results: 43 differences found in the comparisons
DQMHistoTests: Total files compared: 48
DQMHistoTests: Total histograms compared: 3297369
DQMHistoTests: Total failures: 9
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 3297340
DQMHistoTests: Total skipped: 20
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 47 files compared)
Checked 202 log files, 165 edm output root files, 48 DQM output files
TriggerResults: no differences found

GPU Comparison Summary

Summary:

No significant changes to the logs found
Reco comparison results: 48 differences found in the comparisons
DQMHistoTests: Total files compared: 3
DQMHistoTests: Total histograms compared: 39740
DQMHistoTests: Total failures: 1149
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 38591
DQMHistoTests: Total skipped: 0
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 2 files compared)
Checked 8 log files, 10 edm output root files, 3 DQM output files
TriggerResults: no differences found

fwyzard · 2024-03-21T16:27:03Z

+heterogeneous

Further improvements will be done in a separate PR.

cmsbuild · 2024-03-21T16:27:29Z

This pull request is fully signed and it will be integrated in one of the next master IBs (tests are also fine). This pull request will now be reviewed by the release team before it's merged. @sextonkennedy, @antoniovilela, @rappoccio (and backports should be raised in the release meeting by the corresponding L2)

antoniovilela · 2024-03-21T17:18:56Z

+1

commit b3970a8 Author: Artur Lobanov <[email protected]> Date: Fri Mar 22 17:21:00 2024 +0100 Cleaning up AXO model from L1Global commit ccb11e8 Author: Artur Lobanov <[email protected]> Date: Fri Mar 22 12:31:05 2024 +0100 Remove spurious comments commit 261c747 Author: Artur Lobanov <[email protected]> Date: Fri Mar 22 12:30:04 2024 +0100 Apply suggestions from code review Co-authored-by: Marco Musich <[email protected]> commit 1530330 Author: Artur Lobanov <[email protected]> Date: Fri Mar 22 11:10:42 2024 +0100 Remove AXO model in customize HLT commit c691834 Author: Artur Lobanov <[email protected]> Date: Fri Mar 22 10:37:44 2024 +0100 Remove debug print commit bbd270a Author: Artur Lobanov <[email protected]> Date: Thu Mar 21 23:36:47 2024 +0100 Apply code format commit 215cb70 Author: Artur Lobanov <[email protected]> Date: Thu Mar 21 23:29:08 2024 +0100 Set AXO condition model from template commit 1dec93d Author: Artur Lobanov <[email protected]> Date: Thu Mar 21 23:29:08 2024 +0100 Set AXO condition model from template commit 3e82bc4 Author: Artur Lobanov <[email protected]> Date: Thu Mar 21 23:28:48 2024 +0100 Remove Axo model setting from L1 Global commit 2c51218 Author: Artur Lobanov <[email protected]> Date: Thu Mar 21 22:26:54 2024 +0100 Add backwards compatibility to AXO condition in TM parser commit 12c3ddb Author: Artur Lobanov <[email protected]> Date: Fri Mar 22 10:17:34 2024 +0100 Original AXO emu changes from Melissa quinnanm commit 4ce99ab Merge: 7911d8c 233a0ae Author: cmsbuild <[email protected]> Date: Thu Mar 21 20:29:32 2024 +0100 Merge pull request cms-sw#44402 from kyungminparkdrums/fifoflagchange_141X Change FEStatus flag labels in ECAL DQM commit 7911d8c Merge: 86d37c3 70494ae Author: cmsbuild <[email protected]> Date: Thu Mar 21 18:19:23 2024 +0100 Merge pull request cms-sw#44458 from fwyzard/improve_alpaka_Pixel_memory_usage_141x Improve the memory usage in the alpaka pixel reconstruction commit 70494ae Author: Andrea Bocci <[email protected]> Date: Tue Mar 19 18:43:46 2024 +0100 Update OneToManyAssocRandomAccess to use the preferred warp size commit a7047fd Author: Andrea Bocci <[email protected]> Date: Mon Mar 18 23:01:38 2024 +0100 Use cached memory buffers in WordFedAppender Co-authored-by: Matti Kortelainen <[email protected]> commit 5a2feb1 Author: Andrea Bocci <[email protected]> Date: Mon Mar 18 22:41:49 2024 +0100 Improve memory usage in SiPixelRawToCluster::acquire() Allocate device memory only when actually used. Co-authored-by: Matti Kortelainen <[email protected]> commit 4bf2fb9 Author: Andrea Bocci <[email protected]> Date: Mon Mar 18 22:29:46 2024 +0100 Remove unused variable from SiPixelRawToCluster::acquire() Co-authored-by: Matti Kortelainen <[email protected]> commit 418f6e2 Author: Andrea Bocci <[email protected]> Date: Mon Mar 18 22:24:17 2024 +0100 Fix the TrackingRecHitsDevice constructor Fix the hitsModuleStart copy to use the correct source (device-to-device instead of host-to-device). Improve the offsetBPIX2 copy and use the data member as source, to guarantee the lifetime of the source. Remove the queue synchronisation. Update the unit test accordingly. commit 233a0ae Author: Kyungmin Park <[email protected]> Date: Thu Mar 14 13:02:17 2024 +0100 Change FIFOFULL flag to FORCEDFULLSUPP and FULLL1ADESYNC to FIFOFULL Co-authored-by: Andrea Bocci <[email protected]> Co-authored-by: Artur Lobanov <[email protected]> Co-authored-by: cmsbuild <[email protected]> Co-authored-by: Kyungmin Park <[email protected]> Co-authored-by: Marco Musich <[email protected]> Co-authored-by: Matti Kortelainen <[email protected]>

cmsbuild added this to the CMSSW_14_1_X milestone Mar 18, 2024

cmsbuild added reconstruction-pending pending-signatures orp-pending tests-started bug-fix code-checks-pending heterogeneous-pending tracking trk labels Mar 18, 2024

fwyzard mentioned this pull request Mar 18, 2024

Improve the memory usage in the alpaka pixel reconstruction [14.0.x] #44459

Merged

cmsbuild added code-checks-approved and removed code-checks-pending labels Mar 18, 2024

fwyzard and others added 4 commits March 19, 2024 00:26

Remove unused variable from SiPixelRawToCluster::acquire()

4bf2fb9

Co-authored-by: Matti Kortelainen <[email protected]>

Improve memory usage in SiPixelRawToCluster::acquire()

5a2feb1

Allocate device memory only when actually used. Co-authored-by: Matti Kortelainen <[email protected]>

Use cached memory buffers in WordFedAppender

a7047fd

Co-authored-by: Matti Kortelainen <[email protected]>

fwyzard force-pushed the improve_alpaka_Pixel_memory_usage_141x branch from 55b7938 to a7047fd Compare March 18, 2024 23:26

cmsbuild added tests-pending code-checks-pending and removed tests-started code-checks-approved labels Mar 18, 2024

cmsbuild removed the tests-pending label Mar 18, 2024

Update OneToManyAssocRandomAccess to use the preferred warp size

70494ae

cmsbuild added tests-pending code-checks-pending and removed tests-approved code-checks-approved labels Mar 20, 2024

cmsbuild added tests-started and removed tests-pending labels Mar 20, 2024

cmsbuild added code-checks-approved and removed code-checks-pending labels Mar 20, 2024

cmsbuild mentioned this pull request Mar 20, 2024

Alpaka vs CUDA DQM compare modules for pixel tracks objects #43964

Closed

cmsbuild added tests-approved and removed tests-started labels Mar 20, 2024

cmsbuild added fully-signed heterogeneous-approved and removed pending-signatures heterogeneous-pending labels Mar 21, 2024

cmsbuild added orp-approved and removed orp-pending labels Mar 21, 2024

cmsbuild merged commit 7911d8c into cms-sw:master Mar 21, 2024
14 checks passed

This was referenced Mar 22, 2024

[CPP20][ALCA] Replace some enums with constexpr ints #44509

Merged

[CPP20][GEOMETRY] Replace some enums with constexpr ints #44514

Merged

fwyzard deleted the improve_alpaka_Pixel_memory_usage_141x branch April 7, 2024 10:02

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Improve the memory usage in the alpaka pixel reconstruction #44458

Improve the memory usage in the alpaka pixel reconstruction #44458

fwyzard commented Mar 18, 2024 •

edited

Loading

fwyzard commented Mar 18, 2024

fwyzard commented Mar 18, 2024

fwyzard commented Mar 18, 2024

cmsbuild commented Mar 18, 2024 •

edited

Loading

cmsbuild commented Mar 18, 2024

cmsbuild commented Mar 18, 2024

fwyzard commented Mar 18, 2024

fwyzard commented Mar 20, 2024

cmsbuild commented Mar 20, 2024

cmsbuild commented Mar 20, 2024

cmsbuild commented Mar 20, 2024

fwyzard commented Mar 21, 2024

cmsbuild commented Mar 21, 2024

antoniovilela commented Mar 21, 2024

Improve the memory usage in the alpaka pixel reconstruction #44458

Improve the memory usage in the alpaka pixel reconstruction #44458

Conversation

fwyzard commented Mar 18, 2024 • edited Loading

PR description:

PR validation:

Backport status

fwyzard commented Mar 18, 2024

fwyzard commented Mar 18, 2024

fwyzard commented Mar 18, 2024

cmsbuild commented Mar 18, 2024 • edited Loading

cmsbuild commented Mar 18, 2024

cmsbuild commented Mar 18, 2024

fwyzard commented Mar 18, 2024

fwyzard commented Mar 20, 2024

cmsbuild commented Mar 20, 2024

cmsbuild commented Mar 20, 2024

cmsbuild commented Mar 20, 2024

Comparison Summary

GPU Comparison Summary

fwyzard commented Mar 21, 2024

cmsbuild commented Mar 21, 2024

antoniovilela commented Mar 21, 2024

fwyzard commented Mar 18, 2024 •

edited

Loading

cmsbuild commented Mar 18, 2024 •

edited

Loading