Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/28532
Title: | Improving ODE Integration on Graphics Processing Units by Reducing Thread Divergence | Authors: | KOVAC, Thomas HABER, Tom VAN REETH, Frank HENS, Niel |
Issue Date: | 2019 | Publisher: | SPRINGER INTERNATIONAL PUBLISHING AG | Source: | Computational Science – ICCS 2019. Lecture Notes in Computer Science, vol 11587, p. 450-456. | Series/Report: | Lecture notes in computer science | Series/Report no.: | 11538 | Abstract: | Ordinary differential equations are widely used for the mathematical modeling of complex systems in biology and statistics. Since the analysis of such models needs to be performed using numerical integration, many applications can be gravely limited by the computational cost. This paper present a general-purpose integrator that runs massively parallel on graphics processing units. By minimizing thread divergence and bundling similar tasks using linear regression, execution time can be reduced by 40–80% when compared to a naive GPU implementation. Compared to a 36-core CPU implementation, a 150 fold runtime improvement is measured. | Keywords: | Pharmacometrics;Epidemiology;Parallelism;High-Performance Computing;Graphics Processing Units | Document URI: | http://hdl.handle.net/1942/28532 | ISBN: | 9783030227432 | DOI: | 10.1007/978-3-030-22744-9_35 | ISI #: | WOS:000589293800035 | Rights: | Springer Nature Switzerland AG 2019 | Category: | C1 | Type: | Proceedings Paper | Validations: | ecoom 2021 |
Appears in Collections: | Research publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
peer-reviewed-author-version.pdf Restricted Access | Peer-reviewed author version | 503.09 kB | Adobe PDF | View/Open Request a copy |
10.1007@978-3-030-22744-935.pdf Restricted Access | Published version | 407.4 kB | Adobe PDF | View/Open Request a copy |
WEB OF SCIENCETM
Citations
1
checked on Oct 12, 2024
Page view(s)
142
checked on Jul 15, 2022
Download(s)
100
checked on Jul 15, 2022
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.