Non-negative Matrix Factorization on Low-Power Architectures and Accelerators: A Comparative Study. (August 2015)

• Record Type: Journal Article
• Authors: Igual, Francisco D.; García, Carlos; Botella, Guillermo; Piñuel, Luis; Prieto-Matías, Manuel; Tirado, Francisco
• Is Part Of: Computers & electrical engineering. Volume 46(2015)
• Page Start: 139
• Page End: 156
• Publication Date: 2015-08
• Subjects: NMF -- Low-power processors -- GPU -- DSP -- Accelerators; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2015.03.035
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: Performance and energy consumption study of NMF is performed on different systems. General-purpose CPUs yield better execution times at the cost of high energy rates. Low-power architectures offer better trace-off in energy and performance. Abstract: Power consumption is emerging as one of the main concerns in the High Performance Computing (HPC) field. As a growing number of bioinformatics applications require HPC techniques and parallel architectures to meet performance requirements, power consumption arises as an additional limitation when accelerating them. In this paper, we present a comparative study of optimized implementations of the Non-negative Matrix Factorization (NMF), that is widely used in many fields of bioinformatics, taking into account both performance and power consumption. We target a wide range of state-of-the-art parallel architectures, including general-purpose, low-power processors and specific-purpose accelerators like GPUs, DSPs or the Intel Xeon Phi. From our study, we gain insights in both performance and energy consumption for each one of them under a number of experimental conditions, and conclude that the most appropriate architecture is usually a trade-off between performance and energy consumption for a given experimental setup and dataset.