Document Server@UHasselt >
Research >
Research publications >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/23913

Title: Improving Operational Intensity in Data Bound Markov Chain Monte Carlo
Authors: Nemeth, Balazs
Haber, Tom
Ashby, Thomas J.
Lamotte, Wim
Issue Date: 2017
Publisher: Elsevier
Citation: Procedia Computer Science, 108(C), p. 2348-2352 (Art N° 205)
Series/Report: Procedia Computer Science
Series/Report no.: 108
Abstract: Typically, parallel algorithms are developed to leverage the processing power of multiple processors simultaneously speeding up overall execution. At the same time, discrepancy between \{DRAM\} bandwidth and microprocessor speed hinders reaching peak performance. This paper explores how operational intensity improves by performing useful computation during otherwise stalled cycles. While the proposed methodology is applicable to a wide variety of parallel algorithms, and at different scales, the concepts are demonstrated in the machine learning context. Performance improvements are shown for Bayesian logistic regression with a Markov chain Monte Carlo sampler, either with multiple chains or with multiple proposals, on a dense data set two orders of magnitude larger than the last level cache on contemporary systems.
URI: http://hdl.handle.net/1942/23913
Link to publication: http://www.sciencedirect.com/science/article/pii/S1877050917305252
DOI: 10.1016/j.procs.2017.05.024
ISSN: 1877-0509
Category: C1
Type: Proceedings Paper
Appears in Collections: Research publications

Files in This Item:

Description SizeFormat
Published version389.25 kBAdobe PDF

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.