Static-Dynamic analysis for Performance and Accuracy of Data Race Detection in MPI One-Sided Programs - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

Static-Dynamic analysis for Performance and Accuracy of Data Race Detection in MPI One-Sided Programs

Résumé

To take advantage of asynchronous communication mechanisms provided by the recent platforms, the Message Passing Interface (MPI) proposes operations based on one-sided communications. These operations enable a better overlap of communications with computations. However, programmers must manage data consistency and synchronization to avoid data races, which may be a daunting task. In this paper, we propose three solutions to improve the performance and the accuracy of an existing data race detection tool in MPI one-sided programs. First, we present a new node-merging algorithm that takes into account non-adjacent memory accesses. Then, we use an alias analysis to reduce the number of load/store instrumented. Finally, we extend our analyses to manage synchronization routines. Our solutions have been implemented in PARCOACH, a MPI verification tool. Experiments on real-life applications show that our contributions lead to a better accuracy, a reduction of the memory usage by a factor up to 4 of the dynamic analysis and a reduction of the overhead at runtime at larger scale.
Fichier principal
Vignette du fichier
samplepaper.pdf (436.49 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04581890 , version 1 (21-05-2024)

Identifiants

  • HAL Id : hal-04581890 , version 1

Citer

Radjasouria Vinayagame, Van Man Nguyen, Marc Sergent, Samuel Thibault, Emmanuelle Saillard. Static-Dynamic analysis for Performance and Accuracy of Data Race Detection in MPI One-Sided Programs. C3PO 2024 - Compiler-assisted Correctness Checking and Performance Optimization for HPC, May 2024, Hambourg, Germany. ⟨hal-04581890⟩

Collections

CNRS INRIA INRIA2
0 Consultations
0 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More