Local Uniform Mesh Refinement on Vector and Parallel Processors
The numerical solution of two and three dimensional partial differential equations (PDEs) by various discretizations is an important and common problem which requires significant computing resources. In fact, the computational requirements of these problems are so great that the straightforward methods are too expensive in both time and memory for any computer, existing or planned. We will discuss one class of algorithms for this problem with particular emphasis on their suitability for vector and parallel computers.
KeywordsLoad Balance Parallel Computer Linear Array Coarse Grid Fine Grid
Unable to display preview. Download preview PDF.
- R. Bank. A Multi-Level Iterative Method for Nonlinear Elliptic Equations, Elliptic Problem Solvers, Academic Press, 1981. pp. 1–16.Google Scholar
- Marsha J. Berger and Joseph Oliger, Adaptive mesh refinement for hyperbolic partial differential equations, Technical Report Manuscript NA-83–02, Stanford University. March 1983.Google Scholar
- Marsha J. Berger and Antony Jameson, Automatic adaptive grid refinement for the Euler equations, Technical Report DOE/ER/03077–202. Courant Mathematics and Computing Laboratory, New York University. October 1983.Google Scholar
- John H. Bolstad, An adaptive finite difference method for hyperbolic systems in one space dimension, Technical Report LBL-13287-rev, Lawrence Berkeley Laboratory, December 1982.Google Scholar
- Dennis Gannon, Self Adaptive Methods for Parabolic Partial Differential Equations, Technical Report UIUCDCS-R-80–1020. Univ. of Illinois. 1980.Google Scholar
- William Gropp, Local uniform mesh refinement for elliptic partial differential equations, Technical Report YALE/DCS/RR-278. Yale University. Department of Computer Science, July 1983.Google Scholar
- William Gropp, Local uniform mesh refinement on loosely-coupled parallel processors, Technical Report YALE/DCS/RR-352. Yale University. Department of Computer Science. December 1984.Google Scholar
- Alexandru Nicolau and Joseph Fisher, Using an oracle to measure parallelism in single instruction stream programs, The 14 th Annual Microprogramming Workshop, ACM and IEEE Computer Society. 1981, pp. 171–182.Google Scholar