SPMD OpenMP versus MPI on a IBM SMP for 3 Kernels of the NAS Benchmarks

Krawezik, Géraud; Alléon, Guillaume; Cappello, Franck

doi:10.1007/3-540-47847-7_39

“…sharing, examples of SPMD-type implementations on systems with a shared memory model also exist [23,24,25].…”

Section: Single Program Multiple Data Computationmentioning

confidence: 99%

High performance computing for linear acoustic wave simulation

Butt¹

2021

Preprint

View full text Add to dashboard Cite

Parallel computing techniques are applied to a linear acoustic wave model to reduce execution time. Three parallel computing models are developed to parallelize computations. The fork-and-join, SPMD and SIMT models define the execution of parallel computations. The precision and efficiency of the linear acoustic wave model are improved through substantial speedups in all implementations. Furthermore, axisymmetric properties of certain acoustic fields lead to a reduction in the spatio-temporal complexity of those acoustic fields by removing redundant computations. The same linear acoustic wave model is also modified and extended to describe wave propagation across multiple media instead of only a single medium. The developed implementations are integrated into a particularly useful package for high performance simulation of two- or three-dimensional linear acoustic elds generated by realistic sources in various fluid media.

show abstract

“…sharing, examples of SPMD-type implementations on systems with a shared memory model also exist [23,24,25].…”

Section: Single Program Multiple Data Computationmentioning

confidence: 99%

High performance computing for linear acoustic wave simulation

Butt¹

2021

Preprint

View full text Add to dashboard Cite

Parallel computing techniques are applied to a linear acoustic wave model to reduce execution time. Three parallel computing models are developed to parallelize computations. The fork-and-join, SPMD and SIMT models define the execution of parallel computations. The precision and efficiency of the linear acoustic wave model are improved through substantial speedups in all implementations. Furthermore, axisymmetric properties of certain acoustic fields lead to a reduction in the spatio-temporal complexity of those acoustic fields by removing redundant computations. The same linear acoustic wave model is also modified and extended to describe wave propagation across multiple media instead of only a single medium. The developed implementations are integrated into a particularly useful package for high performance simulation of two- or three-dimensional linear acoustic elds generated by realistic sources in various fluid media.

show abstract

“…Other recent papers [17,18] conduct a comparison of OpenMP versus MPI on a specific architecture, the IBM SP3 NH2, for a set of NAS benchmark applications (FT, CG, MG). The authors show that an OpenMP-based parallelization strategy, that takes into account data locality issues, i.e., requires a higher MPI-like programming effort, yields best performance.…”

Section: Related Workmentioning

confidence: 99%

Exploiting Fine-Grained Parallelism in the Phylogenetic Likelihood Function with MPI, Pthreads, and OpenMP: A Performance Study

Stamatakis

¹

,

Ott

²

2008

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Emerging multi-and many-core computer architectures pose new challenges with respect to efficient exploitation of parallelism. In addition, it is currently not clear which might be the most appropriate parallel programming paradigm to exploit such architectures, both from the efficiency as well as software engineering point of view. Beyond that, the application of high performance computing techniques and the use of supercomputers will be essential to deal with the explosive accumulation of sequence data. We address these issues via a thorough performance study by example of RAxML, which is a widely used Bioinformatics application for large-scale phylogenetic inference under the Maximum Likelihood criterion. We provide an overview over the respective parallelization strategies with MPI, Pthreads, and OpenMP and assess performance for these approaches on a large variety of parallel architectures. Results indicate that there is no universally best-suited paradigm with respect to efficiency and portability of the ML function. Therefore, we suggest that the ML function should be parallelized with MPI and Pthreads based on software engineering criteria as well as to enforce data locality.

show abstract

SPMD OpenMP versus MPI on a IBM SMP for 3 Kernels of the NAS Benchmarks

Cited by 5 publications

References 2 publications

High performance computing for linear acoustic wave simulation

High performance computing for linear acoustic wave simulation

High performance computing for linear acoustic wave simulation

Exploiting Fine-Grained Parallelism in the Phylogenetic Likelihood Function with MPI, Pthreads, and OpenMP: A Performance Study

Contact Info

Product

Resources

About