PCERE: Fine-grained Parallel Benchmark Decomposition for Scalability Prediction

Mihail Popov 1 Chadi Akel 1 Florent Conti William Jalby 1 Pablo De Oliveira Castro 1, 2
1 UVSQ - Université de Versailles Saint-Quentin-en-Yvelines
2 LI-PaRAD - Laboratoire d'Informatique Parallélisme Réseaux Algorithmes Distribués
Abstract : Evaluating the strong scalability of OpenMP applications is a costly and time-consuming process. It traditionally requires executing the whole application multiple times with different number of threads. We propose the Parallel Codelet Extractor and REplayer (PCERE), a tool to reduce the cost of scalability evaluation. PCERE decomposes applications into small pieces called codelets: each codelet maps to an OpenMP parallel region and can be replayed as a standalone program. To accelerate scalability prediction, PCERE replays codelets while varying the number of threads. Prediction speedup comes from two key ideas. First, the number of invocations during replay can be significantly reduced. Invocations that have the same performance are grouped together and a single representative is replayed. Second, sequential parts of the programs do not need to be replayed for each different thread configuration. PCERE codelets can be captured once and replayed accurately on multiple architectures, enabling cross-architecture parallel performance prediction. We evaluate PCERE on a C version of the NAS 3.0 Parallel Benchmarks (NPB). We achieve an average speed-up of 25 times on evaluating OpenMP applications scalability with an average error of 4.9\% (median error of 1.7\%).
Type de document :
Communication dans un congrès
2015 IEEE International Parallel and Distributed Processing Symposium (IPDPS), May 2015, Hyderabad, India. 〈10.1109/IPDPS.2015.19〉
Liste complète des métadonnées
https://hal-uvsq.archives-ouvertes.fr/hal-01417304
Contributeur : Pablo De Oliveira Castro <>
Soumis le : jeudi 15 décembre 2016 - 15:09:08
Dernière modification le : mercredi 7 mars 2018 - 14:14:28

Identifiants

Collections

UVSQ | UNIV-PARIS-SACLAY | LI-PARAD

Citation

Mihail Popov, Chadi Akel, Florent Conti, William Jalby, Pablo De Oliveira Castro. PCERE: Fine-grained Parallel Benchmark Decomposition for Scalability Prediction. 2015 IEEE International Parallel and Distributed Processing Symposium (IPDPS), May 2015, Hyderabad, India. 〈10.1109/IPDPS.2015.19〉. 〈hal-01417304〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

60

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD