Identifying thread interference with performance variation

Abstract : Understanding the performance of a multi-threaded appli- cation is difficult. The threads interfere when they access the same hardware resource, which slows down their execution. Unfortunately, current profiling tools are unable to identify the most problematic interference, because they cannot classify interference on different hardware resources. In this paper, we propose a holistic metric able to simultaneously classify inter- ference on different hardware resources. The metric considers performance variation as a universal indicator. We propose an interactive profiling tool to compute this metric. With an evaluation of 27 applications we show that our profiling tool successfully identify 10 performance bottlenecks caused by 6 different kinds of interference
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01784556
Contributor : Médiathèque Télécom Sudparis & Institut Mines-Télécom Business School <>
Submitted on : Thursday, May 3, 2018 - 3:44:28 PM
Last modification on : Wednesday, June 12, 2019 - 1:34:46 AM

Identifiers

  • HAL Id : hal-01784556, version 1

Citation

Mohamed Saïd Mosli Bouksiaa, François Trahay, Alexis Lescouet, Gauthier Voron, Amina Guermouche, et al.. Identifying thread interference with performance variation. [Research Report] Dept. Informatique (Institut Mines-Télécom-Télécom SudParis); Services répartis, Architectures, MOdélisation, Validation, Administration des Réseaux (Institut Mines-Télécom-Télécom SudParis-CNRS); Laboratoire Informatique Paris VI (Université Pierre et Marie Curie - Paris VI). 2017, pp.14. ⟨hal-01784556⟩

Share

Metrics

Record views

89