DH-Falcon: A Language for Large-Scale Graph Processing on Distributed Heterogeneous Systems

Graph models of social information systems typically contain trillions of edges. Such big graphs cannot beprocessed on a single machine. The graph object must bepartitioned and distributed among machines and processedin parallel on a computer cluster. Programming such systemsis very challenging. In...

Full description

Saved in:
Bibliographic Details
Main Authors: Cheramangalath, Unnikrishnan, Nasre, Rupesh, Srikant, Y. N.
Format: Conference Proceeding
Language:English
Subjects:
Clustering algorithms
Computational modeling
CUDA
Distribute systems
Distributed graph processing
Domain-specific languages
DSL
Dynamic graph algorithms
Falcon
Graphics processing units
Heuristic algorithms
Mirrors
MPI
OpenMP
Partitioning algorithms
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 450
container_issue
container_start_page 439
container_title
container_volume
creator Cheramangalath, Unnikrishnan
Nasre, Rupesh
Srikant, Y. N.
description Graph models of social information systems typically contain trillions of edges. Such big graphs cannot beprocessed on a single machine. The graph object must bepartitioned and distributed among machines and processedin parallel on a computer cluster. Programming such systemsis very challenging. In this work, we present DH-Falcon, a graph DSL (domain-specific language) which can be usedto implement parallel algorithms for large-scale graphs, tar-geting Distributed Heterogeneous (CPU and GPU) clusters. DH-Falcon compiler is built on top of the Falcon compiler, which targets single node devices with CPU and multipleGPUs. An important facility provided by DH-Falcon is that itsupports mutation of graph objects, which allows programmerto write dynamic graph algorithms. Experimental evaluationshows that DH-Falcon matches or outperforms state-of-the-art frameworks and gains a speedup of up to 13×.
doi_str_mv 10.1109/CLUSTER.2017.72
format conference_proceeding
fullrecord <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_8048957</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>8048957</ieee_id><sourcerecordid>8048957</sourcerecordid><originalsourceid>FETCH-LOGICAL-i175t-572e98529ba47ab2f747a99e82623f49b2ab92e4946688f9907fad224849e3643</originalsourceid><addsrcrecordid>eNotjj1PwzAUAA0SEqV0ZmDxH0ixnx3bj63qJ1IkEGknhspJX0JQm1R2MvTfUwmm0y2nY-xJiqmUAl_m2S7fLj-nIKSdWrhhE7ROpsoZUGDcLRuBNC5BSNU9e4jxRwhllTAj9rXYJCt_LLv2lc945tt68DXxqgtXCTUleemPxNfBn7_5R-hKirFpa961fNHEPjTF0NOBb6in0NXUUjdEnl9iT6f4yO4qf4w0-eeY7VbL7XyTZO_rt_ksSxpp0z5JLRC6FLDw2voCKnsFIjm43lcaC_AFAmnUxjhXIQpb-QOAdhpJGa3G7Pmv2xDR_hyakw-XvRPaYWrVL1xvUaA</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>DH-Falcon: A Language for Large-Scale Graph Processing on Distributed Heterogeneous Systems</title><source>IEEE Xplore All Conference Series</source><creator>Cheramangalath, Unnikrishnan ; Nasre, Rupesh ; Srikant, Y. N.</creator><creatorcontrib>Cheramangalath, Unnikrishnan ; Nasre, Rupesh ; Srikant, Y. N.</creatorcontrib><description>Graph models of social information systems typically contain trillions of edges. Such big graphs cannot beprocessed on a single machine. The graph object must bepartitioned and distributed among machines and processedin parallel on a computer cluster. Programming such systemsis very challenging. In this work, we present DH-Falcon, a graph DSL (domain-specific language) which can be usedto implement parallel algorithms for large-scale graphs, tar-geting Distributed Heterogeneous (CPU and GPU) clusters. DH-Falcon compiler is built on top of the Falcon compiler, which targets single node devices with CPU and multipleGPUs. An important facility provided by DH-Falcon is that itsupports mutation of graph objects, which allows programmerto write dynamic graph algorithms. Experimental evaluationshows that DH-Falcon matches or outperforms state-of-the-art frameworks and gains a speedup of up to 13×.</description><identifier>EISSN: 2168-9253</identifier><identifier>EISBN: 9781538623268</identifier><identifier>EISBN: 1538623269</identifier><identifier>DOI: 10.1109/CLUSTER.2017.72</identifier><identifier>CODEN: IEEPAD</identifier><language>eng</language><publisher>IEEE</publisher><subject>Clustering algorithms ; Computational modeling ; CUDA ; Distribute systems ; Distributed graph processing ; Domain-specific languages ; DSL ; Dynamic graph algorithms ; Falcon ; Graphics processing units ; Heuristic algorithms ; Mirrors ; MPI ; OpenMP ; Partitioning algorithms</subject><ispartof>2017 IEEE International Conference on Cluster Computing (CLUSTER), 2017, p.439-450</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/8048957$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,27925,54555,54932</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/8048957$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Cheramangalath, Unnikrishnan</creatorcontrib><creatorcontrib>Nasre, Rupesh</creatorcontrib><creatorcontrib>Srikant, Y. N.</creatorcontrib><title>DH-Falcon: A Language for Large-Scale Graph Processing on Distributed Heterogeneous Systems</title><title>2017 IEEE International Conference on Cluster Computing (CLUSTER)</title><addtitle>CLUSTER</addtitle><description>Graph models of social information systems typically contain trillions of edges. Such big graphs cannot beprocessed on a single machine. The graph object must bepartitioned and distributed among machines and processedin parallel on a computer cluster. Programming such systemsis very challenging. In this work, we present DH-Falcon, a graph DSL (domain-specific language) which can be usedto implement parallel algorithms for large-scale graphs, tar-geting Distributed Heterogeneous (CPU and GPU) clusters. DH-Falcon compiler is built on top of the Falcon compiler, which targets single node devices with CPU and multipleGPUs. An important facility provided by DH-Falcon is that itsupports mutation of graph objects, which allows programmerto write dynamic graph algorithms. Experimental evaluationshows that DH-Falcon matches or outperforms state-of-the-art frameworks and gains a speedup of up to 13×.</description><subject>Clustering algorithms</subject><subject>Computational modeling</subject><subject>CUDA</subject><subject>Distribute systems</subject><subject>Distributed graph processing</subject><subject>Domain-specific languages</subject><subject>DSL</subject><subject>Dynamic graph algorithms</subject><subject>Falcon</subject><subject>Graphics processing units</subject><subject>Heuristic algorithms</subject><subject>Mirrors</subject><subject>MPI</subject><subject>OpenMP</subject><subject>Partitioning algorithms</subject><issn>2168-9253</issn><isbn>9781538623268</isbn><isbn>1538623269</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2017</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotjj1PwzAUAA0SEqV0ZmDxH0ixnx3bj63qJ1IkEGknhspJX0JQm1R2MvTfUwmm0y2nY-xJiqmUAl_m2S7fLj-nIKSdWrhhE7ROpsoZUGDcLRuBNC5BSNU9e4jxRwhllTAj9rXYJCt_LLv2lc945tt68DXxqgtXCTUleemPxNfBn7_5R-hKirFpa961fNHEPjTF0NOBb6in0NXUUjdEnl9iT6f4yO4qf4w0-eeY7VbL7XyTZO_rt_ksSxpp0z5JLRC6FLDw2voCKnsFIjm43lcaC_AFAmnUxjhXIQpb-QOAdhpJGa3G7Pmv2xDR_hyakw-XvRPaYWrVL1xvUaA</recordid><startdate>201709</startdate><enddate>201709</enddate><creator>Cheramangalath, Unnikrishnan</creator><creator>Nasre, Rupesh</creator><creator>Srikant, Y. N.</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>201709</creationdate><title>DH-Falcon: A Language for Large-Scale Graph Processing on Distributed Heterogeneous Systems</title><author>Cheramangalath, Unnikrishnan ; Nasre, Rupesh ; Srikant, Y. N.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i175t-572e98529ba47ab2f747a99e82623f49b2ab92e4946688f9907fad224849e3643</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2017</creationdate><topic>Clustering algorithms</topic><topic>Computational modeling</topic><topic>CUDA</topic><topic>Distribute systems</topic><topic>Distributed graph processing</topic><topic>Domain-specific languages</topic><topic>DSL</topic><topic>Dynamic graph algorithms</topic><topic>Falcon</topic><topic>Graphics processing units</topic><topic>Heuristic algorithms</topic><topic>Mirrors</topic><topic>MPI</topic><topic>OpenMP</topic><topic>Partitioning algorithms</topic><toplevel>online_resources</toplevel><creatorcontrib>Cheramangalath, Unnikrishnan</creatorcontrib><creatorcontrib>Nasre, Rupesh</creatorcontrib><creatorcontrib>Srikant, Y. N.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE/IET Electronic Library</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Cheramangalath, Unnikrishnan</au><au>Nasre, Rupesh</au><au>Srikant, Y. N.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>DH-Falcon: A Language for Large-Scale Graph Processing on Distributed Heterogeneous Systems</atitle><btitle>2017 IEEE International Conference on Cluster Computing (CLUSTER)</btitle><stitle>CLUSTER</stitle><date>2017-09</date><risdate>2017</risdate><spage>439</spage><epage>450</epage><pages>439-450</pages><eissn>2168-9253</eissn><eisbn>9781538623268</eisbn><eisbn>1538623269</eisbn><coden>IEEPAD</coden><abstract>Graph models of social information systems typically contain trillions of edges. Such big graphs cannot beprocessed on a single machine. The graph object must bepartitioned and distributed among machines and processedin parallel on a computer cluster. Programming such systemsis very challenging. In this work, we present DH-Falcon, a graph DSL (domain-specific language) which can be usedto implement parallel algorithms for large-scale graphs, tar-geting Distributed Heterogeneous (CPU and GPU) clusters. DH-Falcon compiler is built on top of the Falcon compiler, which targets single node devices with CPU and multipleGPUs. An important facility provided by DH-Falcon is that itsupports mutation of graph objects, which allows programmerto write dynamic graph algorithms. Experimental evaluationshows that DH-Falcon matches or outperforms state-of-the-art frameworks and gains a speedup of up to 13×.</abstract><pub>IEEE</pub><doi>10.1109/CLUSTER.2017.72</doi><tpages>12</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier EISSN: 2168-9253
ispartof 2017 IEEE International Conference on Cluster Computing (CLUSTER), 2017, p.439-450
issn 2168-9253
language eng
recordid cdi_ieee_primary_8048957
source IEEE Xplore All Conference Series
subjects Clustering algorithms
Computational modeling
CUDA
Distribute systems
Distributed graph processing
Domain-specific languages
DSL
Dynamic graph algorithms
Falcon
Graphics processing units
Heuristic algorithms
Mirrors
MPI
OpenMP
Partitioning algorithms
title DH-Falcon: A Language for Large-Scale Graph Processing on Distributed Heterogeneous Systems
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-29T06%3A00%3A19IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=DH-Falcon:%20A%20Language%20for%20Large-Scale%20Graph%20Processing%20on%20Distributed%20Heterogeneous%20Systems&rft.btitle=2017%20IEEE%20International%20Conference%20on%20Cluster%20Computing%20(CLUSTER)&rft.au=Cheramangalath,%20Unnikrishnan&rft.date=2017-09&rft.spage=439&rft.epage=450&rft.pages=439-450&rft.eissn=2168-9253&rft.coden=IEEPAD&rft_id=info:doi/10.1109/CLUSTER.2017.72&rft.eisbn=9781538623268&rft.eisbn_list=1538623269&rft_dat=%3Cieee_CHZPO%3E8048957%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i175t-572e98529ba47ab2f747a99e82623f49b2ab92e4946688f9907fad224849e3643%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=8048957&rfr_iscdi=true