Fredrik Berg Kjolstad
Marc Snir
ABSTRACT
Many problems consist of a structured grid of points that are updated repeatedly based on the values of a fixed set of neighboring points in the same grid. To parallelize these problems we can geometrically divide the grid into chunks that are processed by different processors. One challenge with this approach is that the update of points at the periphery of a chunk requires values from neighboring chunks. These are often located in remote memory belonging to different processes. The naive implementation results in a lot of time spent on communication leaving less time for useful computation. By using the Ghost Cell Pattern communication overhead can be reduced. This results in faster time to completion.
- The message passing interface (mpi) standard. http://www.mcs.anl.gov/research/projects/mpi/, Retrieved June 14 2010.Google Scholar
- Petsc -- portable, extensible toolkit for scientific computation. http://www.mcs.anl.gov/petsc/petsc-as/index.html, Retrieved June 14 2010.Google Scholar
- N. Chen, R. K. Karmani, A. Shali, B.-Y. Su, and R. Johnson. Collective communication patterns. http://parlab.eecs.berkeley.edu/wiki/_media/patterns/paraplop_g1_4.pdf, April 30 2009.Google Scholar
- W. Gropp and E. Lusk. Tuning mpi programs for peak performance. http://www.mcs.anl.gov/research/projects/mpi/tutorials/perf/, 1997.Google Scholar
- E.-G. Kim and M. Snir. Wavefront pattern. http://www.cs.illinois.edu/~snir/PPP/patterns/wavefront.pdf, Retrieved June 14 2010.Google Scholar
- T. Mattson and K. Keutzer. Our pattern language. http://parlab.eecs.berkeley.edu/wiki/patterns/patterns, Retrieved June 14 2010.Google Scholar
- B. Wilkinson and M. Allen. Parallel Programming, Techniques and Applications Using Networked Workstations and Parallel Computers. Pearson Prentice Hall, 2005. Google ScholarDigital Library
Index Terms
- Ghost Cell Pattern
Recommendations
A ghost cell expansion method for reducing communications in solving PDE problems
SC '01: Proceedings of the 2001 ACM/IEEE conference on SupercomputingIn solving Partial Differential Equations, such as the Barotropic equations in ocean models, on Distributed Memory Computers, finite difference methods are commonly used. Most often, processor subdomain boundaries must be updated at each time step. This ...
Parallel and Streaming Generation of Ghost Data for Structured Grids
Parallel simulations decompose large domains into many blocks. A fundamental requirement for subsequent parallel analysis and visualization is the presence of ghost data that supplements each block with a layer of adjacent data elements from neighboring ...
Embracing Scientific Workflow Knowledge into IDE
SKG '08: Proceedings of the 2008 Fourth International Conference on Semantics, Knowledge and GridScientific workflow is mainly supported in many development and execution environments corresponding to specific domains. Although there are many differences among those environments, some consensuses have been achieved to a certain extent in the ...
Comments