Software infrastructure for non-uniform scientific computations on parallel processors

Non-Uniform representations are an effective means of resolving diverse physical phenomena but they require sophisticated run time support to manage their underlying complexity, especially on parallel computers. I will describe a software infrastructure for non-uniform computations that was developed in my research group, and discuss how this infrastructure was employed in a computational grand challenge: first principle simulation of real materials. At the heart of our infrastructure are the LPARX and KeLP systems, C++ class libraries that offer portable performance across a diversity of hardware platforms.

[1]  P. Colella,et al.  Local adaptive mesh refinement for shock hydrodynamics , 1989 .

[2]  Scott B. Baden,et al.  Analysis of the numerical effects of parallelism on a parallel genetic algorithm , 1996, Proceedings of International Conference on Parallel Processing.

[3]  Barbara M. Chapman,et al.  Extending HPF for Advanced Data-Parallel Applications , 1994, IEEE Parallel & Distributed Technology: Systems & Applications.

[4]  Scott B. Baden,et al.  Irregular Coarse-Grain Data Parallelism under LPARX , 1996, Sci. Program..

[5]  Joel H. Saltz,et al.  An Integrated Runtime and Compile-Time Approach for Parallelizing Structured and Block Structured Applications , 1995, IEEE Trans. Parallel Distributed Syst..

[6]  Piyush Mehrotra,et al.  Dynamic data distributions in Vienna Fortran , 1993, Supercomputing '93.

[7]  William L. Briggs,et al.  A multigrid tutorial , 1987 .

[8]  Scott B. Baden,et al.  A parallel software infrastructure for dynamic block-irregular scientific calculations , 1995 .

[9]  Rice UniversityCORPORATE,et al.  High performance Fortran language specification , 1993 .

[10]  Scott B. Baden,et al.  Scalable Parallel Numerical Methods and Software Tools for Material Design , 1994, PPSC.

[11]  Scott B. Baden,et al.  Parallel Cluster Identification for Multidimensional Lattices , 1997, IEEE Trans. Parallel Distributed Syst..

[12]  Scott B. Baden,et al.  Flexible Communication Mechanisms for Dynamic Structured Applications , 1996, IRREGULAR.

[13]  M. Berger,et al.  Adaptive mesh refinement for hyperbolic partial differential equations , 1982 .