No Cover Image

Journal article 1464 views 41 downloads

InK-Compact: In-Kernel Stream Compaction and Its Application to Multi-Kernel Data Visualization on General-Purpose GPUs

D. M Hughes, I. S Lim, M. W Jones, A Knoll, B Spencer, Mark Jones Orcid Logo

Computer Graphics Forum, Volume: 32, Issue: 6, Pages: 178 - 188

Swansea University Author: Mark Jones Orcid Logo

Check full text

DOI (Published version): 10.1111/cgf.12083

Abstract

Stream compaction is an important parallel computing primitive that produces a reduced (compacted) output stream consisting of only valid elements from an input stream containing both invalid and valid elements. Computing on this compacted stream rather than the mixed input stream leads to improveme...

Full description

Published in: Computer Graphics Forum
ISSN: 0167-7055
Published: 2013
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa15061
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2013-07-23T12:13:46Z
last_indexed 2022-06-15T02:25:08Z
id cronfa15061
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2022-06-14T15:52:47.2561609</datestamp><bib-version>v2</bib-version><id>15061</id><entry>2013-06-13</entry><title>InK-Compact: In-Kernel Stream Compaction and Its Application to Multi-Kernel Data Visualization on General-Purpose GPUs</title><swanseaauthors><author><sid>2e1030b6e14fc9debd5d5ae7cc335562</sid><ORCID>0000-0001-8991-1190</ORCID><firstname>Mark</firstname><surname>Jones</surname><name>Mark Jones</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2013-06-13</date><deptcode>SCS</deptcode><abstract>Stream compaction is an important parallel computing primitive that produces a reduced (compacted) output stream consisting of only valid elements from an input stream containing both invalid and valid elements. Computing on this compacted stream rather than the mixed input stream leads to improvements in performance, load balancing and memory footprint. Stream compaction has numerous applications in a wide range of domains: e.g. deferred shading, isosurface extraction and surface voxelization in computer graphics and visualization. We present a novel In-Kernel stream compaction method, where compaction is completed before leaving an operating kernel. This contrasts with conventional parallel compaction methods that require leaving the kernel and running a prefix sum kernel followed by a scatter kernel. We apply our compaction methods to ray-tracing-based visualization of volumetric data. We demonstrate that the proposed In-Kernel compaction outperforms the standard out-of-kernel Thrust parallel-scan method for performing stream compaction in this real-world application. For the data visualization, we also propose a novel multi-kernel ray-tracing pipeline for increased thread coherency and show that it outperforms a conventional single-kernel approach.</abstract><type>Journal Article</type><journal>Computer Graphics Forum</journal><volume>32</volume><journalNumber>6</journalNumber><paginationStart>178</paginationStart><paginationEnd>188</paginationEnd><publisher/><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>0167-7055</issnPrint><issnElectronic/><keywords/><publishedDay>31</publishedDay><publishedMonth>12</publishedMonth><publishedYear>2013</publishedYear><publishedDate>2013-12-31</publishedDate><doi>10.1111/cgf.12083</doi><url/><notes/><college>COLLEGE NANME</college><department>Computer Science</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>SCS</DepartmentCode><institution>Swansea University</institution><apcterm>Not Required</apcterm><lastEdited>2022-06-14T15:52:47.2561609</lastEdited><Created>2013-06-13T13:56:07.0614533</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>D. M</firstname><surname>Hughes</surname><order>1</order></author><author><firstname>I. S</firstname><surname>Lim</surname><order>2</order></author><author><firstname>M. W</firstname><surname>Jones</surname><order>3</order></author><author><firstname>A</firstname><surname>Knoll</surname><order>4</order></author><author><firstname>B</firstname><surname>Spencer</surname><order>5</order></author><author><firstname>Mark</firstname><surname>Jones</surname><orcid>0000-0001-8991-1190</orcid><order>6</order></author></authors><documents><document><filename>15061__24315__4c137e94173f42fd80dccbb1a5684eb6.pdf</filename><originalFilename>2013_CGF_FK-Compact.pdf</originalFilename><uploaded>2022-06-14T15:52:02.2528698</uploaded><type>Output</type><contentLength>8564355</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><documentNotes>Linking old record to accepted manuscript</documentNotes><copyrightCorrect>false</copyrightCorrect></document></documents><OutputDurs/></rfc1807>
spelling 2022-06-14T15:52:47.2561609 v2 15061 2013-06-13 InK-Compact: In-Kernel Stream Compaction and Its Application to Multi-Kernel Data Visualization on General-Purpose GPUs 2e1030b6e14fc9debd5d5ae7cc335562 0000-0001-8991-1190 Mark Jones Mark Jones true false 2013-06-13 SCS Stream compaction is an important parallel computing primitive that produces a reduced (compacted) output stream consisting of only valid elements from an input stream containing both invalid and valid elements. Computing on this compacted stream rather than the mixed input stream leads to improvements in performance, load balancing and memory footprint. Stream compaction has numerous applications in a wide range of domains: e.g. deferred shading, isosurface extraction and surface voxelization in computer graphics and visualization. We present a novel In-Kernel stream compaction method, where compaction is completed before leaving an operating kernel. This contrasts with conventional parallel compaction methods that require leaving the kernel and running a prefix sum kernel followed by a scatter kernel. We apply our compaction methods to ray-tracing-based visualization of volumetric data. We demonstrate that the proposed In-Kernel compaction outperforms the standard out-of-kernel Thrust parallel-scan method for performing stream compaction in this real-world application. For the data visualization, we also propose a novel multi-kernel ray-tracing pipeline for increased thread coherency and show that it outperforms a conventional single-kernel approach. Journal Article Computer Graphics Forum 32 6 178 188 0167-7055 31 12 2013 2013-12-31 10.1111/cgf.12083 COLLEGE NANME Computer Science COLLEGE CODE SCS Swansea University Not Required 2022-06-14T15:52:47.2561609 2013-06-13T13:56:07.0614533 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science D. M Hughes 1 I. S Lim 2 M. W Jones 3 A Knoll 4 B Spencer 5 Mark Jones 0000-0001-8991-1190 6 15061__24315__4c137e94173f42fd80dccbb1a5684eb6.pdf 2013_CGF_FK-Compact.pdf 2022-06-14T15:52:02.2528698 Output 8564355 application/pdf Accepted Manuscript true Linking old record to accepted manuscript false
title InK-Compact: In-Kernel Stream Compaction and Its Application to Multi-Kernel Data Visualization on General-Purpose GPUs
spellingShingle InK-Compact: In-Kernel Stream Compaction and Its Application to Multi-Kernel Data Visualization on General-Purpose GPUs
Mark Jones
title_short InK-Compact: In-Kernel Stream Compaction and Its Application to Multi-Kernel Data Visualization on General-Purpose GPUs
title_full InK-Compact: In-Kernel Stream Compaction and Its Application to Multi-Kernel Data Visualization on General-Purpose GPUs
title_fullStr InK-Compact: In-Kernel Stream Compaction and Its Application to Multi-Kernel Data Visualization on General-Purpose GPUs
title_full_unstemmed InK-Compact: In-Kernel Stream Compaction and Its Application to Multi-Kernel Data Visualization on General-Purpose GPUs
title_sort InK-Compact: In-Kernel Stream Compaction and Its Application to Multi-Kernel Data Visualization on General-Purpose GPUs
author_id_str_mv 2e1030b6e14fc9debd5d5ae7cc335562
author_id_fullname_str_mv 2e1030b6e14fc9debd5d5ae7cc335562_***_Mark Jones
author Mark Jones
author2 D. M Hughes
I. S Lim
M. W Jones
A Knoll
B Spencer
Mark Jones
format Journal article
container_title Computer Graphics Forum
container_volume 32
container_issue 6
container_start_page 178
publishDate 2013
institution Swansea University
issn 0167-7055
doi_str_mv 10.1111/cgf.12083
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str 1
active_str 0
description Stream compaction is an important parallel computing primitive that produces a reduced (compacted) output stream consisting of only valid elements from an input stream containing both invalid and valid elements. Computing on this compacted stream rather than the mixed input stream leads to improvements in performance, load balancing and memory footprint. Stream compaction has numerous applications in a wide range of domains: e.g. deferred shading, isosurface extraction and surface voxelization in computer graphics and visualization. We present a novel In-Kernel stream compaction method, where compaction is completed before leaving an operating kernel. This contrasts with conventional parallel compaction methods that require leaving the kernel and running a prefix sum kernel followed by a scatter kernel. We apply our compaction methods to ray-tracing-based visualization of volumetric data. We demonstrate that the proposed In-Kernel compaction outperforms the standard out-of-kernel Thrust parallel-scan method for performing stream compaction in this real-world application. For the data visualization, we also propose a novel multi-kernel ray-tracing pipeline for increased thread coherency and show that it outperforms a conventional single-kernel approach.
published_date 2013-12-31T03:17:12Z
_version_ 1763750373866078208
score 11.014067

Similar Items