Results 1 
4 of
4
Parallel Program Archetypes
 In Proceedings of the Scalable Parallel Library Conference
, 1997
"... A parallel program archetype is an abstraction that captures the common features of a class of problems with similar computational structure and combines them with a parallelization strategy to produce a pattern of dataflow and communication. Such abstractions are useful in application developme ..."
Abstract

Cited by 21 (0 self)
 Add to MetaCart
A parallel program archetype is an abstraction that captures the common features of a class of problems with similar computational structure and combines them with a parallelization strategy to produce a pattern of dataflow and communication. Such abstractions are useful in application development, both as a conceptual framework and as a basis for tools and techniques. This paper describes an approach to parallel application development based on archetypes and presents two example archetypes with applications. 1 Introduction This paper proposes a specific method of exploiting computational and dataflow patterns to help in developing reliable parallel programs. A great deal of work has been done on methods of exploiting design patterns in program development. This paper restricts attention to one kind of pattern that is relevant in parallel programming: the pattern of the parallel computation and communication structure. Methods of exploiting design patterns in program develop...
Performance and Portability of an Air Quality Model
 Parallel Computing
, 1997
"... We present a portable, parallel implementation of an urban air quality model. The parallel model runs on the Intel Delta, Intel Paragon, IBM SP2, and Cray T3D, using a variety of standard communication libraries. We analyze the performance of the air quality model on these platforms based on a model ..."
Abstract

Cited by 8 (0 self)
 Add to MetaCart
(Show Context)
We present a portable, parallel implementation of an urban air quality model. The parallel model runs on the Intel Delta, Intel Paragon, IBM SP2, and Cray T3D, using a variety of standard communication libraries. We analyze the performance of the air quality model on these platforms based on a model derived from the parallel communication behavior and sequential execution time of the air quality model. We predict the performance of the next generation air quality models based on this analysis. 1 Introduction Air quality models (AQMs) predict the spatial and temporal distribution of gaseous species concentrations in the atmosphere. Pollution dynamics is governed by a rich set of physical and chemical phenomenon including advection, turbulent diffusion, chemical transformations, emissions, and deposition. AQMs are mainly used for the evaluation of emission control strategies and planning for the control of air pollution episodes. The California Institute of Technology (CIT) photochemical...
A some fast solver for system of reactiondiffusion equations
 Domain Decomposition Methods in Science and Engineering CINME
, 2002
"... In this paper we present a fast algorithm for the numerical solution of systems of reactiondiffusion equations, ∂tu + a · ∇u = ∆u + F(x, t, u), x ∈ Ω ⊂ R 3, t> 0. (1) Here, u is a vectorvalued function, u ≡ u(x, t) ∈ R m, m is large, and the ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
(Show Context)
In this paper we present a fast algorithm for the numerical solution of systems of reactiondiffusion equations, ∂tu + a · ∇u = ∆u + F(x, t, u), x ∈ Ω ⊂ R 3, t> 0. (1) Here, u is a vectorvalued function, u ≡ u(x, t) ∈ R m, m is large, and the
A Grid Solver for ReactionConvectionDiffusion Operators 1
, 2007
"... In this paper, we present and analyze the performance of a fast parallel distributed computing time integration procedure for systems of ReactionConvectionDiffusion equations. Typical applications include large scale computing of air quality models or population models in biology for which the mai ..."
Abstract
 Add to MetaCart
(Show Context)
In this paper, we present and analyze the performance of a fast parallel distributed computing time integration procedure for systems of ReactionConvectionDiffusion equations. Typical applications include large scale computing of air quality models or population models in biology for which the main solver corresponds to a ReactionConvectionDiffusion operator. One starts from a stabilized explicit time stepping scheme analyzed by Dupros et al (Int. Journal for Numerical Methods in Fluids, 2006). The numerical efficiency and the parallel scalability of this algorithm have been demonstrated on homogeneous parallel architectures. We introduce here an additional domain decomposition component to the algorithm to extend the scalability of the method to multicluster architectures. The targeted computer architecture is a high latency/low bandwidth network of few parallel systems. This paper provides one of the rare examples of a Partial Differential Equation application that is both numerically efficient and scalable on a wide area network of O(10) parallel systems.