FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2904192253> ?p ?o ?g. }

• W2904192253 abstract "Graph processing is increasingly bottlenecked by main memory accesses. On-chip caches are of little help because the irregular structure of graphs causes seemingly random memory references. However, most real-world graphs offer significant potential locality—it is just hard to predict ahead of time. In practice, graphs have well-connected regions where relatively few vertices share edges with many common neighbors. If these vertices were processed together, graph processing would enjoy significant data reuse. Hence, a graph's traversal schedule largely determines its locality. This paper explores online traversal scheduling strategies that exploit the community structure of real-world graphs to improve locality. Software graph processing frameworks use simple, locality-oblivious scheduling because, on general-purpose cores, the benefits of locality-aware scheduling are outweighed by its overheads. Software frameworks rely on offline preprocessing to improve locality. Unfortunately, preprocessing is so expensive that its costs often negate any benefits from improved locality. Recent graph processing accelerators have inherited this design. Our insight is that this misses an opportunity: Hardware acceleration allows for more sophisticated, online locality-aware scheduling than can be realized in software, letting systems significantly improve locality without any preprocessing. To exploit this insight, we present bounded depth-first scheduling (BDFS), a simple online locality-aware scheduling strategy. BDFS restricts each core to explore one small, connected region of the graph at a time, improving locality on graphs with good community structure. We then present HATS, a hardware-accelerated traversal scheduler that adds just 0.4% area and 0.2% power over general-purpose cores. We evaluate BDFS and HATS on several algorithms using large real-world graphs. On a simulated 16-core system, BDFS reduces main memory accesses by up to 2.4x and by 30% on average. However, BDFS is too expensive in software and degrades performance by 21% on average. HATS eliminates these overheads, allowing BDFS to improve performance by 83% on average (up to 3.1x) over a locality-oblivious software implementation and by 31% on average (up to 2.1x) over specialized prefetchers." @default.
• W2904192253 created "2018-12-22" @default.
• W2904192253 creator A5008156365 @default.
• W2904192253 creator A5023312480 @default.
• W2904192253 creator A5049205401 @default.
• W2904192253 creator A5061646272 @default.
• W2904192253 creator A5083800790 @default.
• W2904192253 date "2018-10-01" @default.
• W2904192253 modified "2023-09-30" @default.
• W2904192253 title "Exploiting Locality in Graph Analytics through Hardware-Accelerated Traversal Scheduling" @default.
• W2904192253 cites W1783256592 @default.
• W2904192253 cites W1955594754 @default.
• W2904192253 cites W1979870592 @default.
• W2904192253 cites W1981943579 @default.
• W2904192253 cites W2000041758 @default.
• W2904192253 cites W2008620264 @default.
• W2904192253 cites W2016289973 @default.
• W2904192253 cites W2016559894 @default.
• W2904192253 cites W2029577083 @default.
• W2904192253 cites W2034050320 @default.
• W2904192253 cites W2034102265 @default.
• W2904192253 cites W2035080386 @default.
• W2904192253 cites W2045271686 @default.
• W2904192253 cites W2046526098 @default.
• W2904192253 cites W2067354926 @default.
• W2904192253 cites W2072725684 @default.
• W2904192253 cites W2095420020 @default.
• W2904192253 cites W2095875205 @default.
• W2904192253 cites W2101196063 @default.
• W2904192253 cites W2113235308 @default.
• W2904192253 cites W2131717044 @default.
• W2904192253 cites W2134237243 @default.
• W2904192253 cites W2139146018 @default.
• W2904192253 cites W2143676586 @default.
• W2904192253 cites W2151690061 @default.
• W2904192253 cites W2170382128 @default.
• W2904192253 cites W2234355962 @default.
• W2904192253 cites W2282294254 @default.
• W2904192253 cites W2284169075 @default.
• W2904192253 cites W2324320295 @default.
• W2904192253 cites W2444127451 @default.
• W2904192253 cites W2465756754 @default.
• W2904192253 cites W2484446135 @default.
• W2904192253 cites W2518649968 @default.
• W2904192253 cites W2545376626 @default.
• W2904192253 cites W2612654866 @default.
• W2904192253 cites W2730999914 @default.
• W2904192253 cites W2772167387 @default.
• W2904192253 cites W2795118915 @default.
• W2904192253 cites W2962865652 @default.
• W2904192253 cites W2962903741 @default.
• W2904192253 cites W4235101327 @default.
• W2904192253 cites W4241140669 @default.
• W2904192253 cites W4252563462 @default.
• W2904192253 doi "https://doi.org/10.1109/micro.2018.00010" @default.
• W2904192253 hasPublicationYear "2018" @default.
• W2904192253 type Work @default.
• W2904192253 sameAs 2904192253 @default.
• W2904192253 citedByCount "78" @default.
• W2904192253 countsByYear W29041922532019 @default.
• W2904192253 countsByYear W29041922532020 @default.
• W2904192253 countsByYear W29041922532021 @default.
• W2904192253 countsByYear W29041922532022 @default.
• W2904192253 countsByYear W29041922532023 @default.
• W2904192253 crossrefType "proceedings-article" @default.
• W2904192253 hasAuthorship W2904192253A5008156365 @default.
• W2904192253 hasAuthorship W2904192253A5023312480 @default.
• W2904192253 hasAuthorship W2904192253A5049205401 @default.
• W2904192253 hasAuthorship W2904192253A5061646272 @default.
• W2904192253 hasAuthorship W2904192253A5083800790 @default.
• W2904192253 hasBestOaLocation W29041922532 @default.
• W2904192253 hasConcept C11413529 @default.
• W2904192253 hasConcept C115537543 @default.
• W2904192253 hasConcept C120314980 @default.
• W2904192253 hasConcept C126255220 @default.
• W2904192253 hasConcept C132525143 @default.
• W2904192253 hasConcept C138885662 @default.
• W2904192253 hasConcept C140745168 @default.
• W2904192253 hasConcept C165696696 @default.
• W2904192253 hasConcept C173608175 @default.
• W2904192253 hasConcept C199360897 @default.
• W2904192253 hasConcept C206729178 @default.
• W2904192253 hasConcept C27602214 @default.
• W2904192253 hasConcept C2779808786 @default.
• W2904192253 hasConcept C33923547 @default.
• W2904192253 hasConcept C34736171 @default.
• W2904192253 hasConcept C38652104 @default.
• W2904192253 hasConcept C41008148 @default.
• W2904192253 hasConcept C41895202 @default.
• W2904192253 hasConcept C80444323 @default.
• W2904192253 hasConcept C96333769 @default.
• W2904192253 hasConceptScore W2904192253C11413529 @default.
• W2904192253 hasConceptScore W2904192253C115537543 @default.
• W2904192253 hasConceptScore W2904192253C120314980 @default.
• W2904192253 hasConceptScore W2904192253C126255220 @default.
• W2904192253 hasConceptScore W2904192253C132525143 @default.
• W2904192253 hasConceptScore W2904192253C138885662 @default.
• W2904192253 hasConceptScore W2904192253C140745168 @default.
• W2904192253 hasConceptScore W2904192253C165696696 @default.
• W2904192253 hasConceptScore W2904192253C173608175 @default.

• next