No Cover Image

Journal article 1135 views 166 downloads

Implementing generalized deep-copy in MPI

Joss Whittle, Rita Borgo Orcid Logo, Mark Jones Orcid Logo

PeerJ Computer Science, Volume: 2, Start page: e95

Swansea University Authors: Rita Borgo Orcid Logo, Mark Jones Orcid Logo

  • peerj-cs-95.pdf

    PDF | Version of Record

    Released under the terms of a Creative Commons Attribution License (CC-BY).

    Download (3.99MB)

Check full text

DOI (Published version): 10.7717/peerj-cs.95

Abstract

In this paper we introduce a framework for implementing deep copy on top of MPI. The process is initiated by passing just the root object of the dynamic data structure. Our framework takes care of all pointer traversal, communication, copying and reconstruction on receiving nodes. The benefit of our...

Full description

Published in: PeerJ Computer Science
ISSN: 2376-5992
Published: PeerJ 2016
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa31197
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2016-11-23T14:16:54Z
last_indexed 2020-07-31T18:47:53Z
id cronfa31197
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2020-07-31T17:39:52.1928030</datestamp><bib-version>v2</bib-version><id>31197</id><entry>2016-11-23</entry><title>Implementing generalized deep-copy in MPI</title><swanseaauthors><author><sid>c4675d4072e4b2b3921ae57666f1d9ff</sid><ORCID>0000-0003-2875-6793</ORCID><firstname>Rita</firstname><surname>Borgo</surname><name>Rita Borgo</name><active>true</active><ethesisStudent>false</ethesisStudent></author><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>2016-11-23</date><deptcode>SCS</deptcode><abstract>In this paper we introduce a framework for implementing deep copy on top of MPI. The process is initiated by passing just the root object of the dynamic data structure. Our framework takes care of all pointer traversal, communication, copying and reconstruction on receiving nodes. The benefit of our approach is that MPI users can deep copy complex dynamic data structures without the need to write bespoke communication or serialize / deserialize methods for each object. These methods can present a challenging implementation problem that can quickly become unwieldy to maintain when working with complex structured data. This paper demonstrates our generic implementation, which encapsulates both approaches. We analyze the approach with a variety of structures (trees, graphs (including complete graphs) and rings) and demonstrate that it performs comparably to hand written implementations, using a vastly simplified programming interface. We make the source code available completely as a convenient header file.</abstract><type>Journal Article</type><journal>PeerJ Computer Science</journal><volume>2</volume><paginationStart>e95</paginationStart><publisher>PeerJ</publisher><issnElectronic>2376-5992</issnElectronic><keywords>MPI extension library, Deep copy, Serialization, Marshalling, Dynamic data structures, Deserialization, Unmarshalling</keywords><publishedDay>21</publishedDay><publishedMonth>11</publishedMonth><publishedYear>2016</publishedYear><publishedDate>2016-11-21</publishedDate><doi>10.7717/peerj-cs.95</doi><url/><notes/><college>COLLEGE NANME</college><department>Computer Science</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>SCS</DepartmentCode><institution>Swansea University</institution><apcterm/><lastEdited>2020-07-31T17:39:52.1928030</lastEdited><Created>2016-11-23T12:59:40.6968766</Created><authors><author><firstname>Joss</firstname><surname>Whittle</surname><order>1</order></author><author><firstname>Rita</firstname><surname>Borgo</surname><orcid>0000-0003-2875-6793</orcid><order>2</order></author><author><firstname>Mark</firstname><surname>Jones</surname><orcid>0000-0001-8991-1190</orcid><order>3</order></author></authors><documents><document><filename>0031197-22122016101824.pdf</filename><originalFilename>peerj-cs-95.pdf</originalFilename><uploaded>2016-12-22T10:18:24.2670000</uploaded><type>Output</type><contentLength>5177623</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>Released under the terms of a Creative Commons Attribution License (CC-BY).</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>http://www.creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2020-07-31T17:39:52.1928030 v2 31197 2016-11-23 Implementing generalized deep-copy in MPI c4675d4072e4b2b3921ae57666f1d9ff 0000-0003-2875-6793 Rita Borgo Rita Borgo true false 2e1030b6e14fc9debd5d5ae7cc335562 0000-0001-8991-1190 Mark Jones Mark Jones true false 2016-11-23 SCS In this paper we introduce a framework for implementing deep copy on top of MPI. The process is initiated by passing just the root object of the dynamic data structure. Our framework takes care of all pointer traversal, communication, copying and reconstruction on receiving nodes. The benefit of our approach is that MPI users can deep copy complex dynamic data structures without the need to write bespoke communication or serialize / deserialize methods for each object. These methods can present a challenging implementation problem that can quickly become unwieldy to maintain when working with complex structured data. This paper demonstrates our generic implementation, which encapsulates both approaches. We analyze the approach with a variety of structures (trees, graphs (including complete graphs) and rings) and demonstrate that it performs comparably to hand written implementations, using a vastly simplified programming interface. We make the source code available completely as a convenient header file. Journal Article PeerJ Computer Science 2 e95 PeerJ 2376-5992 MPI extension library, Deep copy, Serialization, Marshalling, Dynamic data structures, Deserialization, Unmarshalling 21 11 2016 2016-11-21 10.7717/peerj-cs.95 COLLEGE NANME Computer Science COLLEGE CODE SCS Swansea University 2020-07-31T17:39:52.1928030 2016-11-23T12:59:40.6968766 Joss Whittle 1 Rita Borgo 0000-0003-2875-6793 2 Mark Jones 0000-0001-8991-1190 3 0031197-22122016101824.pdf peerj-cs-95.pdf 2016-12-22T10:18:24.2670000 Output 5177623 application/pdf Version of Record true Released under the terms of a Creative Commons Attribution License (CC-BY). true eng http://www.creativecommons.org/licenses/by/4.0/
title Implementing generalized deep-copy in MPI
spellingShingle Implementing generalized deep-copy in MPI
Rita Borgo
Mark Jones
title_short Implementing generalized deep-copy in MPI
title_full Implementing generalized deep-copy in MPI
title_fullStr Implementing generalized deep-copy in MPI
title_full_unstemmed Implementing generalized deep-copy in MPI
title_sort Implementing generalized deep-copy in MPI
author_id_str_mv c4675d4072e4b2b3921ae57666f1d9ff
2e1030b6e14fc9debd5d5ae7cc335562
author_id_fullname_str_mv c4675d4072e4b2b3921ae57666f1d9ff_***_Rita Borgo
2e1030b6e14fc9debd5d5ae7cc335562_***_Mark Jones
author Rita Borgo
Mark Jones
author2 Joss Whittle
Rita Borgo
Mark Jones
format Journal article
container_title PeerJ Computer Science
container_volume 2
container_start_page e95
publishDate 2016
institution Swansea University
issn 2376-5992
doi_str_mv 10.7717/peerj-cs.95
publisher PeerJ
document_store_str 1
active_str 0
description In this paper we introduce a framework for implementing deep copy on top of MPI. The process is initiated by passing just the root object of the dynamic data structure. Our framework takes care of all pointer traversal, communication, copying and reconstruction on receiving nodes. The benefit of our approach is that MPI users can deep copy complex dynamic data structures without the need to write bespoke communication or serialize / deserialize methods for each object. These methods can present a challenging implementation problem that can quickly become unwieldy to maintain when working with complex structured data. This paper demonstrates our generic implementation, which encapsulates both approaches. We analyze the approach with a variety of structures (trees, graphs (including complete graphs) and rings) and demonstrate that it performs comparably to hand written implementations, using a vastly simplified programming interface. We make the source code available completely as a convenient header file.
published_date 2016-11-21T03:38:05Z
_version_ 1763751688172208128
score 11.035634

Similar Items