What multilevel parallel programs do when you are not watching: a performance analysis case study comparing MPI/OpenMP, MLP, and nested OpenMP

Authors:
Gabriele Jost;Jesús Labarta;Judit Gimenez
Affiliations:
NAS Division, NASA Ames Research Center, CA;European Center for Parallelism of Barcelona-Technical, University of Catalonia (CEPBA-UPC), Barcelona, Spain;European Center for Parallelism of Barcelona-Technical, University of Catalonia (CEPBA-UPC), Barcelona, Spain
Venue:
WOMPAT'04 Proceedings of the 5th international conference on OpenMP Applications and Tools: shared Memory Parallel Programming with OpenMP
Year:
2004

Citing 5
Cited 1

Thread fork/join techniques for multi-level parallelism exploitation in NUMA multiprocessors

ICS '99 Proceedings of the 13th international conference on Supercomputing
Achieving 60 GFLOP/s on the production CFD code OverFLow-MLP

Parallel Computing - Special issue on parallel computing in aerospace
A comparison of three programming models for adaptive applications on the origin2000

Journal of Parallel and Distributed Computing
A Comparison of MPI, SHMEM and Cache-Coherent Shared Address Space Programming Models on a Tightly-Coupled Multiprocessors

International Journal of Parallel Programming
Performance Analysis of Multilevel Parallel Applications on Shared Memory Architectures

IPDPS '03 Proceedings of the 17th International Symposium on Parallel and Distributed Processing

Execution model of three parallel languages: OpenMP, UPC and CAF

Scientific Programming - International Symposium of Parallel and Distributed Computing & International Workshop on Algorithms, Models and Tools for Parallel Computing on Heterogenous Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we present a performance analysis case study of two multilevel parallel benchmark codes implemented in three different programming paradigms applicable to shared memory computer architectures. We describe how detailed analysis techniques help to differentiate between the influences of the programming model itself and other factors, such as implementation specific behavior of the operating system or architectural issues.