Synchronization-Avoiding Graph Algorithms
Because they were developed for optimal sequential complexity, classical graph algorithms as found in textbooks have strictly-defined orders of operations. Enforcing a prescribed order of operations, or even an approximate order, in a distributed memory setting requires significant amounts of synchr...
Uložené v:
| Vydané v: | 2018 IEEE 25th International Conference on High Performance Computing (HiPC) s. 52 - 61 |
|---|---|
| Hlavní autori: | , , , |
| Médium: | Konferenčný príspevok.. |
| Jazyk: | English |
| Vydavateľské údaje: |
IEEE
01.12.2018
|
| Predmet: | |
| ISSN: | 2640-0316 |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | Because they were developed for optimal sequential complexity, classical graph algorithms as found in textbooks have strictly-defined orders of operations. Enforcing a prescribed order of operations, or even an approximate order, in a distributed memory setting requires significant amounts of synchronization, which in turn can severely limit scalability. As a result, new algorithms are typically required to achieve scalable performance, even for solving well-known graph problems. Yet, even in these cases, parallel graph algorithms are written according to parallel programming models that evolved for, e.g., scientific computing, and that still have inherent, and scalability-limiting, amounts of synchronization. In this paper we present a new approach to parallel graph algorithms: synchronization-avoiding algorithms. To eliminate synchronization and its associated overhead, synchronization-avoiding algorithms perform work in an unordered and fully asynchronous fashion in such a way that the result is constantly refined toward its final state. "Wasted" work is minimized by locally prioritizing tasks using problem-dependent task utility metrics. We classify algorithms for graph applications into two broad categories: algorithms with monotonic updates (which evince global synchronization) and algorithms with non-monotonic updates (which evince vertex-centric synchronization). We apply our approach to both classes and develop novel, synchronization-avoiding algorithms for solving exemplar problems: SSSP and connected components for the former, graph coloring for the latter. We demonstrate that eliminating synchronization in conjunction with effective scheduling policies and optimizations in the runtime results in improved scalability for both classes of algorithms. |
|---|---|
| AbstractList | Because they were developed for optimal sequential complexity, classical graph algorithms as found in textbooks have strictly-defined orders of operations. Enforcing a prescribed order of operations, or even an approximate order, in a distributed memory setting requires significant amounts of synchronization, which in turn can severely limit scalability. As a result, new algorithms are typically required to achieve scalable performance, even for solving well-known graph problems. Yet, even in these cases, parallel graph algorithms are written according to parallel programming models that evolved for, e.g., scientific computing, and that still have inherent, and scalability-limiting, amounts of synchronization. In this paper we present a new approach to parallel graph algorithms: synchronization-avoiding algorithms. To eliminate synchronization and its associated overhead, synchronization-avoiding algorithms perform work in an unordered and fully asynchronous fashion in such a way that the result is constantly refined toward its final state. "Wasted" work is minimized by locally prioritizing tasks using problem-dependent task utility metrics. We classify algorithms for graph applications into two broad categories: algorithms with monotonic updates (which evince global synchronization) and algorithms with non-monotonic updates (which evince vertex-centric synchronization). We apply our approach to both classes and develop novel, synchronization-avoiding algorithms for solving exemplar problems: SSSP and connected components for the former, graph coloring for the latter. We demonstrate that eliminating synchronization in conjunction with effective scheduling policies and optimizations in the runtime results in improved scalability for both classes of algorithms. |
| Author | Firoz, Jesun Sahariar Kanewala, Thejaka Zalewski, Marcin Lumsdaine, Andrew |
| Author_xml | – sequence: 1 givenname: Jesun Sahariar surname: Firoz fullname: Firoz, Jesun Sahariar – sequence: 2 givenname: Marcin surname: Zalewski fullname: Zalewski, Marcin – sequence: 3 givenname: Thejaka surname: Kanewala fullname: Kanewala, Thejaka – sequence: 4 givenname: Andrew surname: Lumsdaine fullname: Lumsdaine, Andrew |
| BookMark | eNotjcFLwzAYxaMoOGfPHrzs6qHdl3xJmh5LcZswcDDvI02-rpEtHW0R5l9vQQ-Pd_g9fu-R3cUuEmPPHDLOoVhuwq7KBHCTAQBXNywpcsMVGm2m6Fs2E1pCCsj1A0uG4WuaiYlpzGfsdX-Nru27GH7sGLqYlt9d8CEeF-veXtpFeTp2fRjb8_DE7ht7Gij57znbr94-q026_Vi_V-U2DQLEmMrau0KjA97UNaITjaqNAU7Tv6JcGeWtdNZSTo30EoRAh4qsb5wnj3P28mcNRHS49OFs--vBaDRcSvwFWWhD-A |
| CODEN | IEEPAD |
| ContentType | Conference Proceeding |
| DBID | 6IE 6IL CBEJK RIE RIL |
| DOI | 10.1109/HiPC.2018.00015 |
| DatabaseName | IEEE Electronic Library (IEL) Conference Proceedings IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume IEEE Xplore All Conference Proceedings IEEE Electronic Library (IEL) IEEE Proceedings Order Plans (POP All) 1998-Present |
| DatabaseTitleList | |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Computer Science |
| EISBN | 9781538683866 1538683865 |
| EISSN | 2640-0316 |
| EndPage | 61 |
| ExternalDocumentID | 8638144 |
| Genre | orig-research |
| GroupedDBID | 29H 29O 6IE 6IF 6IL 6IN AAWTH ABLEC ADZIZ ALMA_UNASSIGNED_HOLDINGS BEFXN BFFAM BGNUA BKEBE BPEOZ CBEJK CHZPO IEGSK OCL RIE RIL |
| ID | FETCH-LOGICAL-i202t-4bdc963c01fbb33c2f5b8801e0315e7585da4caae7ef4d40223c35eadfcded3 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 7 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000465758800006&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| IngestDate | Wed Aug 27 03:02:01 EDT 2025 |
| IsPeerReviewed | false |
| IsScholarly | true |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-i202t-4bdc963c01fbb33c2f5b8801e0315e7585da4caae7ef4d40223c35eadfcded3 |
| PageCount | 10 |
| ParticipantIDs | ieee_primary_8638144 |
| PublicationCentury | 2000 |
| PublicationDate | 2018-Dec |
| PublicationDateYYYYMMDD | 2018-12-01 |
| PublicationDate_xml | – month: 12 year: 2018 text: 2018-Dec |
| PublicationDecade | 2010 |
| PublicationTitle | 2018 IEEE 25th International Conference on High Performance Computing (HiPC) |
| PublicationTitleAbbrev | HIPC |
| PublicationYear | 2018 |
| Publisher | IEEE |
| Publisher_xml | – name: IEEE |
| SSID | ssj0002683637 ssj0020125 |
| Score | 2.108026 |
| Snippet | Because they were developed for optimal sequential complexity, classical graph algorithms as found in textbooks have strictly-defined orders of operations.... |
| SourceID | ieee |
| SourceType | Publisher |
| StartPage | 52 |
| SubjectTerms | Approximation algorithms Asynchronous execution Classification algorithms Color Distributed graph algorithms Image color analysis Runtime Synchronization Task analysis |
| Title | Synchronization-Avoiding Graph Algorithms |
| URI | https://ieeexplore.ieee.org/document/8638144 |
| WOSCitedRecordID | wos000465758800006&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwED61FQNTgRZBeSgDCxKmcZI28VhVlE5VpTJ0q-zzGSJBgtq0Ev8eOwmlAwubZcmSH3c-f-e7-wDugoSQ-9owYbEGi0jHTEZKMGtbrDUljmQqsol4NkuWSzFvwMM-F4aIyuAzenTN8i9f57h1rrJ-YoXFAoAmNON4WOVq7f0pwTAJhwdgy168g7qUD_dFf5rOxy6Qy0VO-o4D94BLpTQlk_b_JnEC3d-cPG--tzan0KDsDNo_pAxeraMduF98ZVhWvK0SLNlol6duiPfsalN7o_fXfJ0Wbx-bLiwmTy_jKav5EFga-EHBIqXR6gv63CgVhhiYgbLqx8kxNZB7-GsZoZQUk4m0BYZBiOHAiopBTTo8h1aWZ3QBXiKNrwKhyYKvSJIU5FwgJBJEg4rzS-i4da8-q4IXq3rJvb-7r-DYbWwV43ENrWK9pRs4wl2Rbta35Sl9A82Mk60 |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LTwIxEJ4gmugJFYxv9-DFxMputwu7R0JEjEhI4MCN9DGVTXTXwELiv7fdXZGDF29NkyZ9zHT6TWfmA7ilIUrPVZpEBmsQhqpNOBMRMbbFWFP0JOqCbKI9HIbTaTSqwP0mFwYR8-AzfLDN_C9fpXJlXWXN0AiLAQA7sBswRt0iW2vjUaGt0G9twS1z9QZlMR_PjZr9eNS1oVw2dtK1LLhbbCq5MenV_jeNQ2j8ZuU5o429OYIKJsdQ-6FlcEotrcPd-CuRec3bIsWSdNZpbIc4T7Y6tdN5f0sXcTb_WDZg3HucdPukZEQgMXVpRphQ0miMdD0thO9LqgNhFNBDy9WA9umvOJOcYxs1UwYaUl_6gREWLRUq_wSqSZrgKTgh166gkUIDvxhHHqF1gmAUSqml8LwzqNt1zz6Lkhezcsnnf3ffwH5_8jqYDZ6HLxdwYDe5iPi4hGq2WOEV7Ml1Fi8X1_mJfQOtkpb0 |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2018+IEEE+25th+International+Conference+on+High+Performance+Computing+%28HiPC%29&rft.atitle=Synchronization-Avoiding+Graph+Algorithms&rft.au=Firoz%2C+Jesun+Sahariar&rft.au=Zalewski%2C+Marcin&rft.au=Kanewala%2C+Thejaka&rft.au=Lumsdaine%2C+Andrew&rft.date=2018-12-01&rft.pub=IEEE&rft.eissn=2640-0316&rft.spage=52&rft.epage=61&rft_id=info:doi/10.1109%2FHiPC.2018.00015&rft.externalDocID=8638144 |