Results 1  10
of
48
The CCA core specification in a distributed memory SPMD framework. Concurrency and Computation: Practice and Experience
"... We present an overview of the CCA core specification and CCAFFEINE, a Sandia National Laboratories framework implementation compliant with the draft specification. CCAFFEINE stands for CCA Fast Framework Example In Need of Everything; that is, CCAFFEINE is fast, lightweight, and it aims to provide e ..."
Abstract

Cited by 54 (13 self)
 Add to MetaCart
(Show Context)
We present an overview of the CCA core specification and CCAFFEINE, a Sandia National Laboratories framework implementation compliant with the draft specification. CCAFFEINE stands for CCA Fast Framework Example In Need of Everything; that is, CCAFFEINE is fast, lightweight, and it aims to provide every "framework service " by using external, portable components instead of integrating all services into a single, heavy framework core. By fast, we mean that the CCAFFEINE glue does not get between components in a way that slows down their interactions. We present the CCAFFEINE solutions to several fundamental problems in the application of component software approaches to the construction of SPMD applications. We demonstrate the integration of components from three
Construction and Application of an AMR Algorithm for Distributed Memory Computers
 PROC. OF CHICAGO WORKSHOP ON ADAPTIVE MESH REFINEMENT METHODS
, 2003
"... ... In this paper, a localitypreserving domain decomposition is proposed that partitions the entire AMR hierarchy from the base level on. It is shown that the approach reduces the communication costs and simplifies the implementation. Emphasis is put on the effective parallelization of the flux cor ..."
Abstract

Cited by 32 (14 self)
 Add to MetaCart
... In this paper, a localitypreserving domain decomposition is proposed that partitions the entire AMR hierarchy from the base level on. It is shown that the approach reduces the communication costs and simplifies the implementation. Emphasis is put on the effective parallelization of the flux correction procedure at coarsefine boundaries, which is indispensable for conservative finite volume schemes. An easily reproducible standard benchmark and a highly resolved parallel AMR simulation of a diffrracting hydrogenoxygen detonation demonstrate the proposed strategy in practice.
A Component Architecture for HighPerformance Computing
 In Proceedings of the Workshop on Performance Optimization via HighLevel Languages and Libraries (POHLL02
, 2002
"... The Common Component Architecture (CCA) provides a means for developers to manage the complexity of largescale scientific software systems and to move toward a "plug and play" environment for highperformance computing. The CCA model allows for a direct connection between components wit ..."
Abstract

Cited by 25 (5 self)
 Add to MetaCart
The Common Component Architecture (CCA) provides a means for developers to manage the complexity of largescale scientific software systems and to move toward a "plug and play" environment for highperformance computing. The CCA model allows for a direct connection between components within the same process to maintain performance on intercomponent calls. It is neutral with respect to parallelism, allowing components to use whatever means they desire to communicate within their parallel "cohort." We will discuss in detail the importance of performance in the design of the CCA and will analyze the performance costs associated with features of the CCA.
A Common Data Management Infrastructure for Adaptive Algorithms for PDE Solutions
 in: Proceedings of the Supercomputing Conference, ACM/IEEE Computer Society
, 1997
"... This paper presents the design, development and application of a computational infrastructure to support the implementation of parallel adaptive algorithms for the solution of sets of partial differential equations. The infrastructure is separated into multiple layers of abstraction. This paper is p ..."
Abstract

Cited by 23 (6 self)
 Add to MetaCart
(Show Context)
This paper presents the design, development and application of a computational infrastructure to support the implementation of parallel adaptive algorithms for the solution of sets of partial differential equations. The infrastructure is separated into multiple layers of abstraction. This paper is primarily concerned with the two lowest layersof this infrastructure: a layer which defines and implements dynamic distributed arrays (DDA), and a layer in which several dynamic data and programming abstractions are implemented in terms of the DDAs. The currently implemented abstractions are those needed for formulation of hierarchical adaptive finite difference methods, hpadaptive finite element methods, and fast multipole method for solution of linear systems. Implementation of sample applications based on each of these methods are described and implementation issues and performance measurements are presented. Keywords:
A Fast Solution Method For ThreeDimensional ManyParticle Problems Of Linear Elasticity
 Int. J. Num. Meth. Engrg
, 1998
"... A boundary element method for solving threedimensional linear elasticity problems that involve a large number of particles embedded in a binder is introduced. The proposed method relies on an iterative solution strategy in which matrixvector multiplication is performed with the fast multipole meth ..."
Abstract

Cited by 20 (6 self)
 Add to MetaCart
A boundary element method for solving threedimensional linear elasticity problems that involve a large number of particles embedded in a binder is introduced. The proposed method relies on an iterative solution strategy in which matrixvector multiplication is performed with the fast multipole method. As a result the method is capable of solving problems with N unknowns using only O(N) memory and O(N) operations. Results are given for problems with hundreds of particles in which N = O(10 5 ). KEY WORDS: boundary element method; fast multipole method; manyparticle problem; linear elasticity; iterative solution strategy . yto whom correspondence should be addressed, email gjr@ticam.utexas.edu 1 Introduction In this paper, we introduce a fast boundary element method (BEM) for solving threedimensional linear elasticity problems that involve a large number of particles embedded in a binder. We refer to those problems as manyparticle problems and to the new method as FLEMS: Fast Li...
HPJava: data parallel extensions to Java
 IN ACM WORKSHOP ON JAVA FOR HIGHPERFORMANCE NETWORK COMPUTING
, 1997
"... We outline an extension of Java for programming with distributed arrays. The basic programming style is Single Program Multiple Data (SPMD), but parallel arrays are provided as new language primitives. Further extensions include three distributed control constructs, the most important being a da ..."
Abstract

Cited by 16 (2 self)
 Add to MetaCart
(Show Context)
We outline an extension of Java for programming with distributed arrays. The basic programming style is Single Program Multiple Data (SPMD), but parallel arrays are provided as new language primitives. Further extensions include three distributed control constructs, the most important being a dataparallel loop construct. Communications involving distributed arrays are handled through a standard library of collective operations. Because the underlying programming model is SPMD programming, direct calls to MPI or other communication packages are also allowed in an HPJava program.
A parallel Vlasov solver using a wavelet based adaptive mesh refinement
 In 7th Workshop on High Perf. Scientific and Engineering Computing (ICPP’2005
, 2005
"... We are interested in solving the Vlasov equation used to describe collective effects in plasmas. This nonlinear partial differential equation coupled with Maxwell equation describes the time evolution of the particle distribution in phase space. The numerical solution of the full threedimensional V ..."
Abstract

Cited by 9 (3 self)
 Add to MetaCart
(Show Context)
We are interested in solving the Vlasov equation used to describe collective effects in plasmas. This nonlinear partial differential equation coupled with Maxwell equation describes the time evolution of the particle distribution in phase space. The numerical solution of the full threedimensional VlasovMaxwell system represents a considerable challenge due to the huge size of the problem. A numerical method based on wavelet transform enables to compute the distribution function on an adaptive mesh from a regular discretization of the phase space. In this paper, we evaluate the costs of this recently developed adaptive scheme applied on a reduced onedimensional model, and its parallelization. We got a fine grain parallel application that achieves a good scalability up to 64 processors on a shared memory architecture. 1
Detonation Structure Simulation with AMROC
 PERFORMANCE COMPUTING AND COMMUNICATIONS 2005
, 2005
"... Numerical simulations can be the key to the thorough understanding of the multidimensional nature of transient detonation waves. But the ..."
Abstract

Cited by 8 (6 self)
 Add to MetaCart
Numerical simulations can be the key to the thorough understanding of the multidimensional nature of transient detonation waves. But the
A numerical study of detonation diffraction
 J. Fluid Mech
, 2004
"... An investigation of detonation diffraction through an abrupt area change has been carried out via a set of twodimensional numerical simulations parameterized by the activation energy of the reactant. Our analysis is specialized to a reactive mixture with a perfect gas equation of state and a single ..."
Abstract

Cited by 7 (3 self)
 Add to MetaCart
An investigation of detonation diffraction through an abrupt area change has been carried out via a set of twodimensional numerical simulations parameterized by the activation energy of the reactant. Our analysis is specialized to a reactive mixture with a perfect gas equation of state and a singlestep reaction in the Arrhenius form. Lagrangian particles are injected into the flow as a diagnostic tool for identifying the dominant terms in the equation that describes the temperature rate of change of a fluid element, expressed in the shockbased reference system. When simplified, this equation provides insight into the competition between the energy release rate and the expansion rate behind the diffracting front. The mechanism of spontaneous generation of transverse waves along the diffracting front is carefully analysed and related to the sensitivity of the reaction rate to temperature. We study in detail three highly resolved cases of detonation diffraction that illustrate different types of behaviour, super, sub and nearcritical diffraction. 1.