Research and development of inefficiency patterns in MPI, UPC applications
Abstract
Most of developed tools for analysis for various libraries (MPI, OpenMP) and languages for parallel programming use low level approaches to analyze the performance of parallel applications. There are a lot of profiling tools and trace visualizers which produce tables, graphs with various statistics of executed program. In most cases developer has to manually look for bottlenecks and opportunities for performance improvement in the produced statistics and graphs. The amount of information developer has to handle manually, increase dramatically with number of cores, number of processes and size of problem in application. Therefore new methods of performance analysis fully or partially handling output information will be more beneficial. To apply the same analysis tool to various parallel paradigm (MPI applications, UPC programs) paradigm-specific inefficiency patterns has been developed. In this paper code patterns resulting in performance penalties are discussed. Patterns of parallel MPI applications for parallel computing systems with distributed memory as well as for parallel UPC programs for systems with partial global address space (PGAS) are considered. A method for automatic detection of inefficiency patterns in parallel MPI applications and UPC programs is proposed. It allows to reduce the tuning time of parallel application.
About the Authors
M. S. Akopyan
ISP RAS, Moscow
Russian Federation
Russian Federation
N. E. Andreev
Thomas Duryea Consulting, Melbourne
Australia
Australia
References
1. Marc Snir, Steve Otto, Steven Huss-Lederman, David Walker, Jack Dongarra. MPI – The complete Reference, Volume 1, The MPI Core, Second edition. The MIT Press. 1998.
2. W. Chen, C. Iancu, K. Yelick. Communication Optimizations for Fine-grained UPC Applications. 14th International Conference on Parallel Architectures and Compilation Techniques (PACT), 2005.
3. Sameer S. Shende Allen D. Malony. “The Tau Parallel Performance System”, International Journal of High Performance Computing Applications, Volume 20 , Issue 2, Pages: 287 – 311, May 2006.
4. L. Li and A.D. Malony, “Model-Based Performance Diagnosis of Master-Worker Parallel Computations,” Lecture Notes in Computer Science, Number 4128, Pages 35-46, 2006.
5. H. Su, M. Billingsley III, and A. George. "Parallel Performance Wizard: A Performance System for the Analysis of Partitioned Global Address Space Applications," International Journal of High-Performance Computing Applications, Vol. 24, No. 4, Nov. 2010, pp. 485-510.
6. Markus Geimer, Felix Wolf, Brian J. N. Wylie, Erika Ábrahám, Daniel Becker, Bernd Mohr. The Scalasca performance toolset architecture. Concurrency and Computation: Practice and Experience, 22(6):702–719, April 2010.
7. Felix Wolf. Automatic Performance Analysis on Parallel Computers with SMP Nodes. PhD thesis, RWTH Aachen, Forschungszentrum Jülich, February 2003, ISBN 3-00-010003-2.
8. Wolf, F., Mohr, B. Automatic performance analysis of hybrid MPI/OpenMP applications. Journal of Systems Architecture 49(10-11) (2003) 421–439.
9. Wolf, F., Mohr, B., Dongarra, J., Moore, S. Efficient Pattern Search in Large Traces through Successive Refinement. In: Proc. European Conf. on Parallel Computing (Euro-Par, Pisa, Italy), Springer (2004).
10. MPICH. http://www.mpich.org
11. MVAPICH. http://mvapich.cse.ohio-state.edu.
12. OpenMPI. http://www.open-mpi.org.
13. Infiniband. http://www.infinibandta.org.
Review
For citations:
Akopyan M.S., Andreev N.E. Research and development of inefficiency patterns in MPI, UPC applications. Proceedings of the Institute for System Programming of the RAS (Proceedings of ISP RAS). 2013;24. (In Russ.)
Email this article 