Big data in multi-block data analysis: An approach to parallelizing Partial Least Squares Mode B algorithm
Other authors
Publication date
2019-04-29ISSN
2405-8440
Abstract
Partial Least Squares (PLS) Mode B is a multi-block method and a tightly coupled algorithm for estimating structural equation models (SEMs). Describing key aspects of parallel computing, we approach the parallelization of the PLS Mode B algorithm to operate on large distributed data. We show the scalability and performance of the algorithm at a very fine-grained level thanks to the versatility of pbdR, a R-project library for parallel computing. We vary several factors under different data distribution schemes in a supercomputing environment. Shorter elapsed times are obtained for the square-blocking factor 16 × 16 using a grid of processors as square as possible and non-square blocking factors 1000 × 4 and 10000 × 4 using an one-column grid of processors. Depending on the configuration, distributing data in a larger number of cores allows reaching speedups of up to 121 over the CPU implementation. Moreover, we show that SEMs can be estimated with big data sets using current state-of-the-art algorithms for multi-block data analysis.
Document Type
Article
Document version
Published version
Language
English
Subject (CDU)
004 - Computer science and technology. Computing. Data processing
Pages
29 p.
Publisher
Elsevier
Is part of
Heliyon
This item appears in the following Collection(s)
Rights
© L'autor/a
Except where otherwise noted, this item's license is described as http://creativecommons.org/licenses/by-nc-nd/4.0/