Parallel performance optimization of large-scale unstructured data visualization for the earth simulator

ABSTRACT

This paper describes some efficient parallel performance optimization strategies for large-scale unstructured data visualization on SMP cluster machines including the Earth Simulator in Japan. The three-level hybrid parallelization is employed in our implementation, consisting of message passing for inter-SMP node communication, loop directives by OpenMP for intra-SMP node parallelization, and vectorization for each processing element (PE). In order to improve the speedup performance for the hybrid parallelization, some techniques, such as multi-coloring for removing data race and dynamic load repartition for load balancing, are considered. Good visualization images and high parallel performance have been achieved on Hitachi SR8000 for large-scale unstructured datasets, which shows the feasibility and effectiveness of our strategies.

REFERENCES

• 1: Accelerated Strategic Computing Initiative (ASCI) Web Site: http://www.llnl.gov/asci/.
• 2: Earth Simulator Research and Development Center Web Site: http://www.es.jamstec.go.jp/.
• 3: GeoFEM Web Site: http://geofem.tokyo.rist.or.jp/.
• 4: I. Fujishiro, L. Chen, Y. Takeshima, H. Nakamura, Y. Suzuki. Parallel visualization of gigabyte datasets in GeoFEM, Journal of Concurrency and Computation: Practice and Experience (in print).
• 5: MPI Web Site: http://www.mpi.org.
• 6: OpenMP Web Site : http//www.openmp.org.
• 7: Hitachi SR8000 Web Site :http://www.hitachi.co.jp/Prod/comp/hpc/foruser/sr8000/.
• 8: Marc Levoy, Display of Surfaces from Volume Data, IEEE Computer Graphics and Applications, v.8 n.3, p.29-37, May 1988 [doi>10.1109/38.511]
• 9: Issei Fujishiro , Yuji Maeda , Hiroshi Sato , Yuriko Takeshima, Volumetric Data Exploration Using Interval Volume, IEEE Transactions on Visualization and Computer Graphics, v.2 n.2, p.144-155, June 1996 [doi>10.1109/2945.506226]
• 10: Brian Cabral , Leith Casey Leedom, Imaging vector fields using line integral convolution, Proceedings of the 20th annual conference on Computer graphics and interactive techniques, p.263-270, September 1993 [doi>10.1145/166117.166151]
• 11: L. Chen, I. Fujishiro and Y. Suzuki. Comprehensible volume LIC rendering based on 3D significance map. Proceedings of SPIE Conference on Visualization and Data Analysis'02, San Jose, 2002, pp. 142-153.
• 12: Thierry Delmarcelle , Lambertus Hesselink, Visualizing Second-Order Tensor Fields with Hyperstreamlines, IEEE Computer Graphics and Applications, v.13 n.4, p.25-33, July 1993 [doi>10.1109/38.219447]
• 13: Issei Fujishiro , Yuriko Takeshima , Taeko Azuma , Shigeo Takahashi, Volume Data Mining Using 3D Field Topology Analysis, IEEE Computer Graphics and Applications, v.20 n.5, p.46-51, September 2000 [doi>10.1109/38.865879]
• 14: James L. Helman , Lambertus Hesselink, Visualizing Vector Field Topology in Fluid Flows, IEEE Computer Graphics and Applications, v.11 n.3, p.36-46, May 1991 [doi>10.1109/38.79452]
• 15: C. M. Wittenbrink. Survey of parallel volume rendering algorithms. Proceedings of International Conference on Parallel distributed Processing Techniques and Applications, Las Vegas, Nevada, 1998, pp. 1329-1336.
• 16: R. Yagel. Towards real time volume rendering. In Proceedings of GRAPH-ICON, Saint-Petersburg, Russia, 1996, pp. 230-241.
• 17: Kwan-Liu Ma , James S. Painter , Charles D. Hansen , Michael F. Krogh, Parallel Volume Rendering Using Binary-Swap Compositing, IEEE Computer Graphics and Applications, v.14 n.4, p.59-68, July 1994 [doi>10.1109/38.291532]
• 18: Craig M. Wittenbrink , Arun K. Somani, Time and space optimal data parallel volume rendering using permutation warping, Journal of Parallel and Distributed Computing, v.46 n.2, p.148-164, Nov. 1, 1997 [doi>10.1006/jpdc.1997.1386]
• 19: Jose Claudio Teixeira E Silva, Jr. , Arie E. Kaufman, Parallel volume rendering of irregular grids, 1996
• 20: Jane Wilhelms , Allen Van Gelder, Octrees for faster isosurface generation, ACM Transactions on Graphics (TOG), v.11 n.3, p.201-227, July 1992 [doi>10.1145/130881.130882]
• 21: C. R. Ramakrishnan, C. Silva. Optimal processor allocation for sort-last compositing under BSP-tree ordering. SPIE Electronic Imaging, Visual Data Exploration and Analysis IV, 1999.
• 22: Marc Levoy, Efficient ray tracing of volume data, ACM Transactions on Graphics (TOG), v.9 n.3, p.245-261, July 1990 [doi>10.1145/78964.78965]
• 23: Rolf Rabenseifner, Communication Bandwidth of Parallel Programming Models on Hybrid Architectures, Proceedings of the 4th International Symposium on High Performance Computing, p.401-412, May 15-17, 2002
• 24: K. Nakajima and H. Okuda. Parallel iterative solvers for unstructured grids using Directive/MPI hybrid programming model for GeoFEM platform on SMP cluster architectures. Journal of Concurrency and Computation:Practice and Experience (in print).
• 25: K. Minami and H. Okuda Performance optimization of GeoFEM on various computer architecture. GeoFEM Report 2001-006, RIST/Tokyo, 2001.
• 26: Kwan-Liu Ma , Thomas W. Crockett, A scalable parallel cell-projection volume rendering algorithm for three-dimensional unstructured data, Proceedings of the IEEE symposium on Parallel rendering, p.95-ff., October 20-21, 1997, Phoenix, Arizona, United States [doi>10.1145/266638.266664]

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.3 COMPUTER GRAPHICS
      I.3.2 Graphics Systems
          Subjects: Distributed/network graphics

Additional Classification:
  C. Computer Systems Organization
  C.2 COMPUTER-COMMUNICATION NETWORKS
      C.2.4 Distributed Systems
          Subjects: Distributed applications; Client/server

  I. Computing Methodologies
  I.3 COMPUTER GRAPHICS
      I.3.7 Three-Dimensional Graphics and Realism

  J. Computer Applications
  J.2 PHYSICAL SCIENCES AND ENGINEERING