TU Darmstadt / ULB / TUbiblio

Towards compiler-aided correctness checking of adjoint MPI applications

Hück, Alexander ; Protze, Joachim ; Lehr, Jan-Patrick ; Terboven, Christian ; Bischof, Christian ; Müller, Matthias S. (2021)
Towards compiler-aided correctness checking of adjoint MPI applications.
2020 IEEE/ACM 4th International Workshop on Software Correctness for HPC Applications (Correctness). (11.11.2020)
doi: 10.26083/tuprints-00018889
Konferenzveröffentlichung, Zweitveröffentlichung, Preprint

Kurzbeschreibung (Abstract)

Algorithmic Differentiation (AD) is a set of techniques to calculate derivatives of a computer program. In C++, AD typically requires (i) a type change of the built-in double, and (ii) a replacement of all MPI calls with AD-specific implementations. This poses challenges on MPI correctness tools, such as MUST, a dynamic checker, and TypeART, its memory sanitizer extension. In particular, AD impacts (i) memory layouts of the whole code, (ii) requires more memory allocations tracking by TypeART, and (iii) approximately doubles the MPI type checks of MUST due to an AD-specific communication reversal. To address these challenges, we propose a new callback interface for MUST to reduce the number of intercepted MPI calls, and, also, improve the filtering capabilities of TypeART to reduce tracking of temporary allocations for the derivative computation. We evaluate our approach on an AD-enhanced version of CORAL LULESH. In particular, we reduce stack variable tracking from 32 million to 13 thousand. MUST with TypeART and the callback interface reduces the runtime overhead to that of vanilla MUST.

Typ des Eintrags: Konferenzveröffentlichung
Erschienen: 2021
Autor(en): Hück, Alexander ; Protze, Joachim ; Lehr, Jan-Patrick ; Terboven, Christian ; Bischof, Christian ; Müller, Matthias S.
Art des Eintrags: Zweitveröffentlichung
Titel: Towards compiler-aided correctness checking of adjoint MPI applications
Sprache: Englisch
Publikationsjahr: 2021
Ort: Darmstadt
Publikationsdatum der Erstveröffentlichung: 2021
Verlag: IEEE
Buchtitel: Proceedings of Correctness 2020: Fourth International Workshop on Software Correctness for HPC Applications
Veranstaltungstitel: 2020 IEEE/ACM 4th International Workshop on Software Correctness for HPC Applications (Correctness)
Veranstaltungsdatum: 11.11.2020
DOI: 10.26083/tuprints-00018889
URL / URN: https://tuprints.ulb.tu-darmstadt.de/18889
Zugehörige Links:
Herkunft: Zweitveröffentlichung
Kurzbeschreibung (Abstract):

Algorithmic Differentiation (AD) is a set of techniques to calculate derivatives of a computer program. In C++, AD typically requires (i) a type change of the built-in double, and (ii) a replacement of all MPI calls with AD-specific implementations. This poses challenges on MPI correctness tools, such as MUST, a dynamic checker, and TypeART, its memory sanitizer extension. In particular, AD impacts (i) memory layouts of the whole code, (ii) requires more memory allocations tracking by TypeART, and (iii) approximately doubles the MPI type checks of MUST due to an AD-specific communication reversal. To address these challenges, we propose a new callback interface for MUST to reduce the number of intercepted MPI calls, and, also, improve the filtering capabilities of TypeART to reduce tracking of temporary allocations for the derivative computation. We evaluate our approach on an AD-enhanced version of CORAL LULESH. In particular, we reduce stack variable tracking from 32 million to 13 thousand. MUST with TypeART and the callback interface reduces the runtime overhead to that of vanilla MUST.

Status: Preprint
URN: urn:nbn:de:tuda-tuprints-188898
Sachgruppe der Dewey Dezimalklassifikatin (DDC): 000 Allgemeines, Informatik, Informationswissenschaft > 004 Informatik
Fachbereich(e)/-gebiet(e): 20 Fachbereich Informatik
20 Fachbereich Informatik > Scientific Computing
Zentrale Einrichtungen
Zentrale Einrichtungen > Hochschulrechenzentrum (HRZ)
Zentrale Einrichtungen > Hochschulrechenzentrum (HRZ) > Hochleistungsrechner
Hinterlegungsdatum: 12 Jul 2021 07:00
Letzte Änderung: 15 Sep 2021 13:57
PPN:
Zugehörige Links:
Export:
Suche nach Titel in: TUfind oder in Google
Frage zum Eintrag Frage zum Eintrag

Optionen (nur für Redakteure)
Redaktionelle Details anzeigen Redaktionelle Details anzeigen