Results 1 
3 of
3
Parallelizing MCMC for Bayesian Spatiotemporal Geostatistical Models
, 2006
"... When MCMC methods for Bayesian spatiotemporal modeling are applied to large geostatistical problems, challenges arise as a consequence of storage requirements, computing costs, and convergence monitoring. This article describes the parallelization of a reparametrized and marginalized posterior samp ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
When MCMC methods for Bayesian spatiotemporal modeling are applied to large geostatistical problems, challenges arise as a consequence of storage requirements, computing costs, and convergence monitoring. This article describes the parallelization of a reparametrized and marginalized posterior sampling (RAMPS) algorithm, which is carefully designed to generate posterior samples efficiently. The algorithm is implemented using the Parallel Linear Algebra Package (PLAPACK). The scalability of the algorithm is investigated via simulation experiments that are implemented using a cluster with 25 processors. The usefulness of the method is illustrated with an application to sulfur dioxide concentration data from the Air Quality System database of the U.S. Environmental Protection Agency.
Speeding up Multibody Analysis by Parallel Computing
, 2000
"... The paper describes the application of parallel techniques to a multibody multidisciplinary formulation in order to speed up the solution of complex nonlinear analyses. The problem is stated in terms of a system of nonlinear DifferentialAlgebraic Equations (DAE). The parallel solution is obtained b ..."
Abstract
 Add to MetaCart
The paper describes the application of parallel techniques to a multibody multidisciplinary formulation in order to speed up the solution of complex nonlinear analyses. The problem is stated in terms of a system of nonlinear DifferentialAlgebraic Equations (DAE). The parallel solution is obtained by means of a coarsescale substructuring domain decomposition method, which is able to exploit the characteristic quasimonodimensional topology that multibody models usually present. The representation of explicit constraints in form of algebraic equations requires particular care in the treatment of the related unknowns, to avoid local singularity problems. The code has been successfully tested on different computer architectures, such as an SMP HPN 4000 and a cluster of PCs. Special attention has been dedicated to producing a code that will efficiently run on the latter type of parallel machines. The analysis of a nonlinear beam bending is presented as a first test case. The algorithm behavior has been also tested on a more complex system such as a helicopter rotor.
© 2008 Science Publications Calculate Sensitivity Function Using Parallel Algorithm
"... Abstract: Problem statement: To calculate sensitivity functions for a large dimension control system using one processor, it takes huge time to find the unknowns vectors for a linear system, which represents the mathematical model of the physical control system. This study is an attempt to solve the ..."
Abstract
 Add to MetaCart
Abstract: Problem statement: To calculate sensitivity functions for a large dimension control system using one processor, it takes huge time to find the unknowns vectors for a linear system, which represents the mathematical model of the physical control system. This study is an attempt to solve the same problem in parallel to reduce the time factor needed and increase the efficiency. Approach: Calculate in parallel sensitivity function using n1 processors where n is a number of linear equations which can be represented as TX = W, where T is a matrix of size n1xn2, X = T −1 W, is a vector of