Multi-core CPUs, Clusters, and Grid Computing: A Tutorial

Creel, Michael; Goffe, William L.

doi:10.1007/s10614-008-9143-5

Cited by 38 publications

(15 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, when the function that is called in line 31 of the previous listing is computationally intensive, then the communication overhead will be small in relation to the part that is parallelized, and we will observe an improvement from parallelization (see Creel and Goffe, 2008). This is the case for the example in the next section.…”

Section: Approximating πmentioning

confidence: 86%

A Note on Julia and MPI, with Code Examples

Creel

2015

Comput Econ

Self Cite

View full text Add to dashboard Cite

This note explains how MPI may be used with the Julia programming language. An example of a simple Monte Carlo study is presented, with code. The code is intended to serve as a general purpose template for more relevant applications. A second example shows how the template code may be adapted to perform a Monte Carlo study of the properties of an Approximate Bayesian Computing estimator of actual research interest. All of the code is available at https://github. com/mcreel/JuliaMPIMonteCarlo.Keywords: Julia programming language; message passing interface; Monte Carlo; approximate Bayesian computing Julia (http://julialang.org) is a fairly new programming language that is oriented toward scientific computing. Development began in 2009, and the current release version as of May 2015 is 0.3.8. While this low release number indicates that the language is still evolving quickly, nevertheless, it is already stable enough to warrant development of code for research. Julia also has a reasonably complete set of features of interest to economists, such as methods for optimization, probability distributions, data manipulation, plotting, and so forth 1 . Julia is a high level language with syntax that is as readable as that of Matlab or Python, but through use of just in time compilation, the developers aim to provide performance similar to that of C, a compiled language. Julia is a free and open source language, and it runs on all popular operating systems. The combination of all of these features makes the language a very appropriate choice for research and teaching in economics.This note shows how Julia may use the message passing interface (MPI) for parallel computing on multicore computers or clusters of computers. Because MPI is such a widely used method of achieving high performance computing for challenging problems, it is very useful to be able to use this method with Julia, as existing code for other languages can easily be adapted. A major advantage of using MPI is code portability: 1 See http://pkg.julialang.org/ for a list -there is even a preliminary Dynare package.

show abstract

Section: Approximating πmentioning

confidence: 86%

A Note on Julia and MPI, with Code Examples

Creel

2015

Comput Econ

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this sense our paper contributes to the recent stream of the literature on the use of Central Processing Unit (CPU) and Graphics Processing Unit (GPU) parallel computing in econometrics (e.g., see Doornik et al (2002), Swann (2002), Creel (2005), Creel and Goffe (2008), Suchard et al (2010)). …”

Section: Introductionmentioning

confidence: 87%

Embarrassingly parallel sequential Markov-chain Monte Carlo for large sets of time series

Casarin

Craiu

Leisen

2016

Statistics and Its Interface

View full text Add to dashboard Cite

Bayesian computation crucially relies on Markov chain Monte Carlo (MCMC) algorithms. In the case of massive data sets, running the Metropolis-Hastings sampler to draw from the posterior distribution becomes prohibitive due to the large number of likelihood terms that need to be calculated at each iteration. In order to perform Bayesian inference for a large set of time series, we consider an algorithm that combines "divide and conquer" ideas previously used to design MCMC algorithms for big data with a sequential MCMC strategy. The performance of the method is illustrated using a large set of financial data.

show abstract

“…First, most of the current processors include SIMD (Single Instruction Multiple Data) instructions, which provide a limited form of parallelism that can be exploited to obtain relevant improvements [7]. Second, multi-core CPUs are able to solve high-performance applications more efficiently by using parallel computing [9]. Third, GPUs have gained an important role in the area of parallel computing [5,34].…”

Section: Parallelization Approachesmentioning

confidence: 99%

Parallelization strategies for markerless human motion capture

Cano

Yeguas-Bolivar

Muñoz-Salinas

et al. 2014

J Real-Time Image Proc

View full text Add to dashboard Cite

Markerless Motion Capture (MMOCAP) is the problem of determining the pose of a person from images captured by one or several cameras simultaneously without using markers on the subject. Evaluation of the solutions is frequently the most time-consuming task, making most of the proposed methods inapplicable in real-time scenarios. This paper presents an efficient approach to parallelize the evaluation of the solutions in CPUs and GPUs. Our proposal is experimentally compared on six sequences of the HumanEva-I dataset using the CMAES algorithm. Multiple algorithm's configurations were tested to analyze the best trade-off in regard to the accuracy and computing time. The proposed methods obtain speedups of 8× in multi-core CPUs, 30× in a single GPU and up to 110× using 4 GPUs.

show abstract

Multi-core CPUs, Clusters, and Grid Computing: A Tutorial

Abstract: Multi-core, Cluster, OpenMP, MPI, Grid,

Cited by 38 publications

References 19 publications

A Note on Julia and MPI, with Code Examples

A Note on Julia and MPI, with Code Examples

Embarrassingly parallel sequential Markov-chain Monte Carlo for large sets of time series

Parallelization strategies for markerless human motion capture

Contact Info

Product

Resources

About