Parallel Solution of Magnetotelluric Occam Inversion Algorithm Based on Hybrid MPI/OpenMP Model

Article Preview

Abstract:

In order to improve the efficiency of magnetotelluric Occam inversion algorithm (MT Occam), a parallel algorithm is implemented on a hybrid MPI/OpenMP parallel programming model to increase its convergence speed and to decrease the operation time. MT Occam is partitioned to map the task on the parallel model. The parallel algorithm implements the coarse-grained parallelism between computation nodes and fine-grained parallelism between cores within each node. By analyzing the data dependency, the computing tasks are accurately partitioned so as to reduce transmission time. The experimental results show that with the increase of model scale, higher speedup can be obtained. The high efficiency of the parallel partitioning strategy of the model can improve the scalability of the parallel algorithm.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

3751-3754

Citation:

Online since:

August 2014

Authors:

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2014 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

* - Corresponding Author

[1] deGroot-Hedlin C. and Constable S.: Occam's inversion to generate smooth, two-dimensional models from magnetotelluric data. Geophysics. 55 (1990), pp.1613-1624.

DOI: 10.1190/1.1442813

Google Scholar

[2] Wu X. P. and Xu G. M.: Improvement of Occam's inversion for MT data. ACTA GEOPHYSICA SINICA. 41 (1998), pp.547-554.

Google Scholar

[3] Li Y., Hu X. Y., Wu G. J., et al: Parallel computation of 2-D magnetotelluric forward modeling based on MPI. Dizhen Dizhi(Seismology and Geology). 32 (2010), pp.392-401.

Google Scholar

[4] Rabenseifner R, Hager G and Jost G: Hybrid MPI/OpenMP parallel programming on clusters of multi-core SMP nodes. Parallel, Distributed and Network-based Processing, 2009 17th Euromicro International Conference on. IEEE. (2009), pp.427-436.

DOI: 10.1109/pdp.2009.43

Google Scholar

[5] SHAN Y., WU J. and WANG Z.: Hierarchical Parallel Programming Model and Parallelization and Optimization Techniques Based on SMP Cluster. Application Research of Computers. 10(2006), pp.254-260.

Google Scholar

[6] Pacheco, P.: An introduction to parallel programming. Burlington: Elsevier. (2011), pp.250-255.

Google Scholar