Parallel implementation of terrain erosion applied to the surface of Mars

ABSTRACT

We present a parallel version of the algorithm that simulates thermal erosion [7]. We split the input data into the strips that are assigned to different processes and run the erosion in parallel. When this task is finished the processes exchange information about the material transported through their boundaries, update their data, and run new erosion step. We use message passing for process synchronization. First, the information about the material transported through the boundary is saved into files and then messages are sent to the corresponding processes informing that their data is ready. Using files for process synchronization makes the implementation platform independent.The parallel version is stable and runs very well on large data. We have achieved speedup 8.4 on ten CPUs. With small data the method is devoting high effort to communication and the speedup decreases. We have tested this algorithm on the 3D map of martial surface obtained from Mars Global Surveyor [8].

REFERENCES

• 1: Bedrich Benes , Rafael Forsbach, Layered Data Representation for Visual Simulation of Terrain Erosion, Proceedings of the 17th Spring conference on Computer graphics, p.80, April 25-28, 2001
• 2: B. Beneš, I. Mar´k, and P. Slavík. Terrain Erosion Model Based On Rewriting of Matrices. In Proceedings of The Fifth International Conference in Central Europe on Computer Graphics and Visualization, pages 341-351. University of West Bohemia Press, 1997.
• 3: N. Chiba, K. Muraoka, and K. Fujita. An Erosion Model Based On Velocity Fields For The Visual Simulation Of Mountain Scenery. The Journal of Visualization and Computer Animation, 9:185-194, 1997.
• 4: D.D. Eckbert, F.K. Musgrave, P. Prusinkiewicz, J. Stam, and J. Tessendors. Simulating Nature: From Theory To Applications. SIGGRAPH 00 Course Notes, pages 1-213, 2000.
• 5: C. Gaillard, F. Zagolski, and F. Bonn. Modelling Of Human Dimension On Soil Erosion Processes For Remote Sensing Applications. In IEEE Geoescience and Remote Sensing Symposium, volume 4, pages 2137-2139, 1997.
• 6: F. Kenton Musgrave, Towards a Synthetic Universe, IEEE Computer Graphics and Applications, v.19 n.6, p.4-5, November 1999 [doi>10.1109/MCG.1999.799711]
• 7: F. K. Musgrave , C. E. Kolb , R. S. Mace, The synthesis and rendering of eroded fractal terrains, Proceedings of the 16th annual conference on Computer graphics and interactive techniques, p.41-50, July 1989
• 8: NASA. http://ltpwww.gsfc.nasa.gov/tharsis/mola.html.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.3 COMPUTER GRAPHICS
      I.3.1 Hardware architecture
          Subjects: Parallel processing

Additional Classification:
  I. Computing Methodologies
  I.3 COMPUTER GRAPHICS
      I.3.5 Computational Geometry and Object Modeling
          Subjects: Physically based modeling

General Terms:
Algorithms, Performance

Keywords:
Parallel algorithms, Terrain erosion, Volume graphics