Parallel Computing Using Multilevel Hypergraph Partitioning for Three-Dimensional Scramjet


As one of the most effective methods to study hypersonic vehicle, the numerical simulation of scramjet has drawn an ever increasing attention. Nevertheless, the serial simulation can't satisfy the practical needs because of high calculation accuracy and considerable computation time. Meanwhile, the efficiency of parallel simulation using the domain decomposition method or graph partitioning software is low. In this paper, we use multilevel hypergraph partitioning to simulate scramjet engine, and define the vertex weight and hyperedge cost by using the averagely measured time. In order to reduce computation while enhancing parallelism, we propose a concept named computing arrays. Furthermore, overlapping calculation and communication is adopted to decrease the communication time when dealing with parallel boundary and coupled communication. Finally, experimental results on an HPC platform using 1 to 32 processors indicate that our approach is effective and practical in scramjet simulation.

  • Abstract
  • Key Words
  • 1. Introduction
  • 2. Governing Equations and Numerical Methods
  • 3. Parallel Strategies
  • 4. Experimental Evaluation
  • 5. Conclusion and Future Work
  • Acknowledgment
  • References

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In