Article

Using MPI derived datatypes in numerical libraries

Authors:

Jesper Larsson TräffAuthors Info & Claims

EuroMPI'11: Proceedings of the 18th European MPI Users' Group conference on Recent advances in the message passing interface

Pages 29 - 38

Published: 18 September 2011 Publication History

Abstract

By way of example this paper examines the potential of MPI user-defined datatypes for distributed datastructure manipulation in numerical libraries. The three examples, namely gather/scatter of columnwise distributed two dimensional matrices, matrix transposition, and redistribution of doubly cyclically distributed matrices as used in the Elemental dense matrix library, show that distributed data structures can be conveniently expressed with the derived datatype mechanisms of MPI, yielding at the same time worthwhile performance advantages over straight-forward, hand written implementations. Experiments have been performed with on different systems with mpich2 and OpenMPI library implementations. We report results for a SunFire X4100 system with the mvapich2 library. We point out cases where the current MPI collective interfaces do not provide sufficient functionality.

References

[1]

Byna, S., Gropp, W.D., Sun, X.-H., Thakur, R.: Improving the performance of MPI derived datatypes by optimizing memory-access cost. In: IEEE International Conference on Cluster Computing (CLUSTER 2003), pp. 412-419 (2003)

Google Scholar

[2]

Byna, S., Sun, X.-H., Thakur, R., Gropp, W.D.: Automatic memory optimizations for improving MPI derived datatype performance. In: Mohr, B., Träff, J.L., Worringen, J., Dongarra, J. (eds.) PVM/MPI 2006. LNCS, vol. 4192, pp. 238-246. Springer, Heidelberg (2006)

Crossref

Google Scholar

[3]

Hoefler, T., Gottlieb, S.: Parallel zero-copy algorithms for fast fourier transform and conjugate gradient using MPI datatypes. In: Keller, R., Gabriel, E., Resch, M., Dongarra, J. (eds.) EuroMPI 2010. LNCS, vol. 6305, pp. 132-141. Springer, Heidelberg (2010)

Crossref

Google Scholar

[4]

Lu, Q., Wu, J., Panda, D.K., Sadayappan, P.: Applying MPI derived datatypes to the NAS benchmarks: A case study. In: 33rd International Conference on Parallel Processing Workshops (ICPP 2004 Workshops), pp. 538-545 (2004)

Crossref

Google Scholar

[5]

MPI Forum. MPI: A Message-Passing Interface Standard. Version 2.2, September 4 (2009), http://www.mpi-forum.org

Google Scholar

[6]

Poulson, J., Marker, B., Hammond, J.R., Romero, N.A., van de Geijn, R.: Elemental: A new framework for distributed memory dense matrix computations. ACM Transactions on Mathematical Software (2011) (conditionally accepted)

Crossref

Google Scholar

[7]

Renault, É.: Extended MPICC to generate MPI derived datatypes from C datatypes automatically. In: Cappello, F., Herault, T., Dongarra, J. (eds.) PVM/MPI 2007. LNCS, vol. 4757, pp. 307-314. Springer, Heidelberg (2007)

Crossref

Google Scholar

[8]

Renault, É., Parrot, C.: MPI pre-processor: Generating MPI derived datatypes from C datatypes automatically. In: International Conference on Parallel Processing Workshops (ICPP), pp. 248-256 (2006)

Crossref

Google Scholar

[9]

Ross, R.J., Miller, N., Gropp, W.D.: Implementing fast and reusable datatype processing. In: Dongarra, J., Laforenza, D., Orlando, S. (eds.) EuroPVM/MPI 2003. LNCS, vol. 2840, pp. 404-413. Springer, Heidelberg (2003)

Google Scholar

[10]

Sanders, P.: Random permutations on distributed, external and hierarchical memory. Information Processing Letters 67(6), 305-309 (1998)

Crossref

Google Scholar

[11]

Träff, J.L.: A (radical) proposal addressing the non-scalability of the irregular MPI collective interfaces. In: 16th International Workshop on High-level Parallel Programming Models and Supportive Environments (HIPS 2011), International Parallel and Distributed Processing Symposium (IPDPS), page 42 (2011)

Crossref

Google Scholar

[12]

Träff, J.L., Hempel, R., Ritzdorf, H., Zimmermann, F.: Flattening on the fly: Efficient handling of MPI derived datatypes. In: Margalef, T., Dongarra, J., Luque, E. (eds.) PVM/MPI 1999. LNCS, vol. 1697, pp. 109-116. Springer, Heidelberg (1999)

Crossref

Google Scholar

[13]

Wu, J., Wyckoff, P., Panda, D.K.: High performance implementation of MPI derived datatype communication over InfiniBand. In: 18th International Parallel and Distributed Processing Symposium (IPDPS 2004), page 14 (2004)

Google Scholar

Cited By

View all

Xiong QBangalore PSkjellum AHerbordt M(2018)MPI Derived DatatypesProceedings of the 25th European MPI Users' Group Meeting10.1145/3236367.3236378(1-10)Online publication date: 23-Sep-2018
https://dl.acm.org/doi/10.1145/3236367.3236378
Puri SPaudel APrasad S(2018)MPI-Vector-IOProceedings of the 47th International Conference on Parallel Processing10.1145/3225058.3225105(1-11)Online publication date: 13-Aug-2018
https://dl.acm.org/doi/10.1145/3225058.3225105
Träff JDongarra JHolmes DCollis ALarsson Träff JSmith L(2016)A Library for Advanced Datatype ProgrammingProceedings of the 23rd European MPI Users' Group Meeting10.1145/2966884.2966904(98-107)Online publication date: 25-Sep-2016
https://dl.acm.org/doi/10.1145/2966884.2966904
Show More Cited By

Recommendations

MPI Derived Datatypes: Performance and Portability Issues
EuroMPI '18: Proceedings of the 25th European MPI Users' Group Meeting

This paper addresses performance-portability and overall performance issues when derived datatypes are used with four MPI implementations: Open MPI, MPICH, MVAPICH2, and Intel MPI. These comparisons are particularly relevant today since most vendor ...
A case for using MPI's derived datatypes to improve I/O performance
SC '98: Proceedings of the 1998 ACM/IEEE conference on Supercomputing

MPI-IO, the I/O part of the MPI-2 standard, is a promising new interface for parallel I/O. A key feature of MPI-IO is that it allows users to access several noncontiguous pieces of data from a file with a single I/O function call by defining file views ...
Collective operations in NEC's high-performance MPI libraries
IPDPS'06: Proceedings of the 20th international conference on Parallel and distributed processing

We give an overview of the algorithms and implementations in the high-performance MPI libraries MPI/SX and MPI/ES of some of the most important collective operations of MPI (the Message Passing Interface). The infrastructure of MPI/SX makes it easy to ...

Comments

Information & Contributors

Information

Published In

EuroMPI'11: Proceedings of the 18th European MPI Users' Group conference on Recent advances in the message passing interface

September 2011

355 pages

ISBN:9783642244483

Editors:
Yiannis Cotronis
University of Athens, Greece
,
Anthony Danalis
University of Tennessee, Knoxville, TN
,
Dimitrios S. Nikolopoulos
University of Crete, Heraklion, Greece
,
Jack Dongarra
University of Tennessee, Knoxville, TN

Publisher

Springer-Verlag

Berlin, Heidelberg

Publication History

Published: 18 September 2011

Qualifiers

Article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

6
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 01 Sep 2024

Other Metrics

View Author Metrics

Citations

Cited By

View all

Xiong QBangalore PSkjellum AHerbordt M(2018)MPI Derived DatatypesProceedings of the 25th European MPI Users' Group Meeting10.1145/3236367.3236378(1-10)Online publication date: 23-Sep-2018
https://dl.acm.org/doi/10.1145/3236367.3236378
Puri SPaudel APrasad S(2018)MPI-Vector-IOProceedings of the 47th International Conference on Parallel Processing10.1145/3225058.3225105(1-11)Online publication date: 13-Aug-2018
https://dl.acm.org/doi/10.1145/3225058.3225105
Träff JDongarra JHolmes DCollis ALarsson Träff JSmith L(2016)A Library for Advanced Datatype ProgrammingProceedings of the 23rd European MPI Users' Group Meeting10.1145/2966884.2966904(98-107)Online publication date: 25-Sep-2016
https://dl.acm.org/doi/10.1145/2966884.2966904
Träff JRougier ADongarra JIshikawa YHori A(2014)Zero-copy, Hierarchical Gather is not possible with MPI Datatypes and CollectivesProceedings of the 21st European MPI Users' Group Meeting10.1145/2642769.2642772(39-44)Online publication date: 9-Sep-2014
https://dl.acm.org/doi/10.1145/2642769.2642772
Träff JDongarra JIshikawa YHori A(2014)Optimal MPI Datatype Normalization for Vector and Index-block TypesProceedings of the 21st European MPI Users' Group Meeting10.1145/2642769.2642771(33-38)Online publication date: 9-Sep-2014
https://dl.acm.org/doi/10.1145/2642769.2642771
Schneider TKjolstad FHoefler TDongarra JBlas JCarretero J(2013)MPI datatype processing using runtime compilationProceedings of the 20th European MPI Users' Group Meeting10.1145/2488551.2488552(19-24)Online publication date: 15-Sep-2013
https://dl.acm.org/doi/10.1145/2488551.2488552

View Options

View options

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Abstract

References

Cited By

Recommendations

MPI Derived Datatypes: Performance and Portability Issues

A case for using MPI's derived datatypes to improve I/O performance

Collective operations in NEC's high-performance MPI libraries

Comments

Information

Published In

Publisher

Publication History

Qualifiers

Contributors

Other Metrics

Bibliometrics

Article Metrics

Other Metrics

Citations

Cited By

View options

Get Access

Login options

Full Access

Figures

Other

Share

Share this Publication link

Share on social media

Affiliations