001041737 001__ 1041737
001041737 005__ 20250804115251.0
001041737 020__ $$a979-8-3503-5554-3
001041737 0247_ $$2doi$$a10.1109/SCW63240.2024.00192
001041737 0247_ $$2WOS$$aWOS:001451792300156
001041737 037__ $$aFZJ-2025-02413
001041737 041__ $$aEnglish
001041737 1001_ $$0P:(DE-Juel1)188670$$aCorbin, Gregor$$b0$$eCorresponding author$$ufzj
001041737 1112_ $$aSC24-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis$$cAtlanta$$d2024-11-17 - 2024-11-22$$gSC24-W$$wGA
001041737 245__ $$aAre Noise-Resilient Logical Timers Useful for Performance Analysis?
001041737 260__ $$bIEEE$$c2024
001041737 300__ $$a1519 - 1530
001041737 3367_ $$2ORCID$$aCONFERENCE_PAPER
001041737 3367_ $$033$$2EndNote$$aConference Paper
001041737 3367_ $$2BibTeX$$aINPROCEEDINGS
001041737 3367_ $$2DRIVER$$aconferenceObject
001041737 3367_ $$2DataCite$$aOutput Types/Conference Paper
001041737 3367_ $$0PUB:(DE-HGF)8$$2PUB:(DE-HGF)$$aContribution to a conference proceedings$$bcontrib$$mcontrib$$s1746444641_28061
001041737 3367_ $$0PUB:(DE-HGF)7$$2PUB:(DE-HGF)$$aContribution to a book$$mcontb
001041737 520__ $$aIn modern HPC systems, performance measurements are often disturbed by noise. Because repeating measurements to increase confidence in their results is costly, alternative noise-resilient techniques are desirable. Therefore, we implement alogical clock, which does not rely on real-time measurements, in Score-P. We explore several methods to model computational work with the clock increment, counting OpenMP loop iterations, LLVM basic blocks/statements, or hardware counters. We demonstrate the strengths and weaknesses of using logical time stamps in a trace analysis workflow with Score-P and Scalasca, by evaluating the performanceproblems we can find in three MPI+OpenMP mini-apps. By design, logical measurementsreliably show algorithmic issues, such as load imbalance, but cannot capture external aspects of program execution, for example memory contention. In summary, logical-time based measurements are a specialized but valuable addition to theperformance analyst’s toolbox.
001041737 536__ $$0G:(DE-HGF)POF4-5111$$a5111 - Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x0
001041737 536__ $$0G:(GEPRIS)449683531$$aDFG project G:(GEPRIS)449683531 - ExtraNoise – Leistungsanalyse von HPC-Anwendungen in verrauschten Umgebungen (449683531)$$c449683531$$x1
001041737 536__ $$0G:(DE-Juel-1)ATMLPP$$aATMLPP - ATML Parallel Performance (ATMLPP)$$cATMLPP$$x2
001041737 588__ $$aDataset connected to CrossRef Conference
001041737 7001_ $$0P:(DE-Juel1)188664$$aDaoud, Nour$$b1
001041737 7001_ $$0P:(DE-Juel1)132199$$aMohr, Bernd$$b2
001041737 7001_ $$0P:(DE-HGF)0$$aDe Morais, Gustavo$$b3
001041737 7001_ $$0P:(DE-Juel1)132299$$aWolf, Felix$$b4
001041737 770__ $$z979-8-3503-5554-3
001041737 773__ $$a10.1109/SCW63240.2024.00192
001041737 8564_ $$uhttps://juser.fz-juelich.de/record/1041737/files/Are_Noise-Resilient_Logical_Timers_Useful_for_Performance_Analysis.pdf$$yRestricted
001041737 909CO $$ooai:juser.fz-juelich.de:1041737$$pVDB
001041737 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)188670$$aForschungszentrum Jülich$$b0$$kFZJ
001041737 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)188664$$aForschungszentrum Jülich$$b1$$kFZJ
001041737 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)132199$$aForschungszentrum Jülich$$b2$$kFZJ
001041737 9131_ $$0G:(DE-HGF)POF4-511$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5111$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x0
001041737 920__ $$lyes
001041737 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x0
001041737 980__ $$acontrib
001041737 980__ $$aVDB
001041737 980__ $$acontb
001041737 980__ $$aI:(DE-Juel1)JSC-20090406
001041737 980__ $$aUNRESTRICTED