GraphScale: scalable Processing on FPGAs for HBM and large graphs

Recent advances in graph processing on FPGAs promise to alleviate performance bottlenecks with irregular memory access patterns. Such bottlenecks challenge performance for a growing number of important application areas like machine learning and data analytics. While FPGAs denote a promising solutio...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Dann, Jonas (VerfasserIn) , Ritter, Daniel (VerfasserIn) , Fröning, Holger (VerfasserIn)
Dokumenttyp: Article (Journal)
Sprache:Englisch
Veröffentlicht: March 2024
In: ACM transactions on reconfigurable technology and Systems
Year: 2024, Jahrgang: 17, Heft: 2, Pages: 1-23
ISSN:1936-7406
DOI:10.1145/3616497
Online-Zugang:Verlag, kostenfrei, Volltext: https://doi.org/10.1145/3616497
Verlag, kostenfrei, Volltext: https://dl.acm.org/doi/10.1145/3616497
Volltext
Verfasserangaben:Jonas Dann, Daniel Ritter, Holger Fröning

MARC

LEADER 00000caa a2200000 c 4500
001 1911194089
003 DE-627
005 20250716214548.0
007 cr uuu---uuuuu
008 241209s2024 xx |||||o 00| ||eng c
024 7 |a 10.1145/3616497  |2 doi 
035 |a (DE-627)1911194089 
035 |a (DE-599)KXP1911194089 
035 |a (OCoLC)1528014536 
040 |a DE-627  |b ger  |c DE-627  |e rda 
041 |a eng 
084 |a 35  |2 sdnb 
100 1 |a Dann, Jonas  |d 1995-  |e VerfasserIn  |0 (DE-588)1290424977  |0 (DE-627)1846075939  |4 aut 
245 1 0 |a GraphScale  |b scalable Processing on FPGAs for HBM and large graphs  |c Jonas Dann, Daniel Ritter, Holger Fröning 
264 1 |c March 2024 
300 |b Illustrationen 
300 |a 23 
336 |a Text  |b txt  |2 rdacontent 
337 |a Computermedien  |b c  |2 rdamedia 
338 |a Online-Ressource  |b cr  |2 rdacarrier 
500 |a Veröffentlicht: 23. März 2024 
500 |a Gesehen am 09.12.2024 
520 |a Recent advances in graph processing on FPGAs promise to alleviate performance bottlenecks with irregular memory access patterns. Such bottlenecks challenge performance for a growing number of important application areas like machine learning and data analytics. While FPGAs denote a promising solution through flexible memory hierarchies and massive parallelism, we argue that current graph processing accelerators either use the off-chip memory bandwidth inefficiently or do not scale well across memory channels.In this work, we propose GraphScale, a scalable graph processing framework for FPGAs. GraphScale combines multi-channel memory with asynchronous graph processing (i.e., for fast convergence on results) and a compressed graph representation (i.e., for efficient usage of memory bandwidth and reduced memory footprint). GraphScale solves common graph problems like breadth-first search, PageRank, and weakly connected components through modular user-defined functions, a novel two-dimensional partitioning scheme, and a high-performance two-level crossbar design. Additionally, we extend GraphScale to scale to modern high-bandwidth memory (HBM) and reduce partitioning overhead of large graphs with binary packing. 
700 1 |a Ritter, Daniel  |e VerfasserIn  |0 (DE-588)1018081852  |0 (DE-627)679891331  |0 (DE-576)354965255  |4 aut 
700 1 |a Fröning, Holger  |d 1976-  |e VerfasserIn  |0 (DE-588)133209466  |0 (DE-627)538678658  |0 (DE-576)299696189  |4 aut 
773 0 8 |i Enthalten in  |a Association for Computing Machinery  |t ACM transactions on reconfigurable technology and Systems  |d New York, NY : ACM Press, 2008  |g 17(2024), 2, Artikel-ID 22, Seite 1-23  |h Online-Ressource  |w (DE-627)561317879  |w (DE-600)2418475-5  |w (DE-576)278983146  |x 1936-7406  |7 nnas 
773 1 8 |g volume:17  |g year:2024  |g number:2  |g elocationid:22  |g pages:1-23  |g extent:23  |a GraphScale scalable Processing on FPGAs for HBM and large graphs 
856 4 0 |u https://doi.org/10.1145/3616497  |x Verlag  |x Resolving-System  |z kostenfrei  |3 Volltext 
856 4 0 |u https://dl.acm.org/doi/10.1145/3616497  |x Verlag  |z kostenfrei  |3 Volltext 
951 |a AR 
992 |a 20241209 
993 |a Article 
994 |a 2024 
998 |g 133209466  |a Fröning, Holger  |m 133209466:Fröning, Holger  |d 700000  |d 720000  |e 700000PF133209466  |e 720000PF133209466  |k 0/700000/  |k 1/700000/720000/  |p 3  |y j 
999 |a KXP-PPN1911194089  |e 4633322621 
BIB |a Y 
SER |a journal 
JSO |a {"title":[{"title_sort":"GraphScale","title":"GraphScale","subtitle":"scalable Processing on FPGAs for HBM and large graphs"}],"type":{"media":"Online-Ressource","bibl":"article-journal"},"id":{"eki":["1911194089"],"doi":["10.1145/3616497"]},"physDesc":[{"extent":"23 S.","noteIll":"Illustrationen"}],"relHost":[{"origin":[{"publisher":"ACM Press","dateIssuedDisp":"2008-","dateIssuedKey":"2008","publisherPlace":"New York, NY"}],"id":{"zdb":["2418475-5"],"issn":["1936-7406","1936-7414"],"eki":["561317879"]},"physDesc":[{"extent":"Online-Ressource"}],"title":[{"title":"ACM transactions on reconfigurable technology and Systems","subtitle":"TRETS","title_sort":"ACM transactions on reconfigurable technology and Systems"}],"recId":"561317879","name":{"displayForm":["Association for Computing Machinery"]},"titleAlt":[{"title":"Transactions on reconfigurable technology and Systems"},{"title":"TRETS"}],"disp":"Association for Computing MachineryACM transactions on reconfigurable technology and Systems","type":{"media":"Online-Ressource","bibl":"periodical"},"corporate":[{"display":"Association for Computing Machinery","role":"aut"}],"note":["Gesehen am 16.06.20"],"pubHistory":["1.2008-"],"part":{"issue":"2","extent":"23","pages":"1-23","text":"17(2024), 2, Artikel-ID 22, Seite 1-23","volume":"17","year":"2024"},"language":["eng"]}],"origin":[{"dateIssuedDisp":"March 2024","dateIssuedKey":"2024"}],"language":["eng"],"name":{"displayForm":["Jonas Dann, Daniel Ritter, Holger Fröning"]},"recId":"1911194089","person":[{"display":"Dann, Jonas","given":"Jonas","role":"aut","family":"Dann"},{"role":"aut","family":"Ritter","display":"Ritter, Daniel","given":"Daniel"},{"family":"Fröning","role":"aut","given":"Holger","display":"Fröning, Holger"}],"note":["Veröffentlicht: 23. März 2024","Gesehen am 09.12.2024"]} 
SRT |a DANNJONASRGRAPHSCALE2024