Matches in SemOpenAlex for { <https://semopenalex.org/work/W3206136464> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W3206136464 abstract "Graph neural networks (GNNs) have gained increasing popularity among researchers recently and have been employed in many applications. Training GNNs introduce a crucial stage called graph sampling. One of the most important sampling algorithms is Random Walk. However, Random Walk and many of its variants share and suffer from the same performance problem caused by random and fragmented memory access patterns, leading to significant system performance degradation. In this work, we present an efficient graph sampling engine on modern FPGAs integrated with in-package high bandwidth memory (HBM), which brings data closer and faster to the core logic. The hardware walker design is modular and easily scalable for massive parallelism, to fully utilize the available HBM channels. Our design also provides the flexibility to support Random Walk and two of its variants on both homogeneous and heterogeneous graphs. On real-world graph datasets, we achieve a 1.39 × -3.74 × speedup with a 2.42 × -6.69 × higher energy efficiency over highly optimized parallel baselines on a Xeon CPU. We also implement these algorithms on a NVIDIA Tesla VIOO GPU and achieve comparable dynamic energy consumption." @default.
- W3206136464 created "2021-10-25" @default.
- W3206136464 creator A5004837462 @default.
- W3206136464 creator A5035198517 @default.
- W3206136464 creator A5046414484 @default.
- W3206136464 creator A5067437853 @default.
- W3206136464 creator A5076699095 @default.
- W3206136464 creator A5076915400 @default.
- W3206136464 creator A5083700947 @default.
- W3206136464 date "2021-08-01" @default.
- W3206136464 modified "2023-10-01" @default.
- W3206136464 title "Graph Sampling with Fast Random Walker on HBM-enabled FPGA Accelerators" @default.
- W3206136464 cites W1977661221 @default.
- W3206136464 cites W2022322548 @default.
- W3206136464 cites W2146591355 @default.
- W3206136464 cites W2513566807 @default.
- W3206136464 cites W2536999580 @default.
- W3206136464 cites W2743104969 @default.
- W3206136464 cites W2962756421 @default.
- W3206136464 cites W2965857891 @default.
- W3206136464 cites W2970929262 @default.
- W3206136464 cites W2982231843 @default.
- W3206136464 cites W3034855459 @default.
- W3206136464 cites W3093557412 @default.
- W3206136464 cites W3104097132 @default.
- W3206136464 cites W3133347161 @default.
- W3206136464 cites W3133395503 @default.
- W3206136464 doi "https://doi.org/10.1109/fpl53798.2021.00042" @default.
- W3206136464 hasPublicationYear "2021" @default.
- W3206136464 type Work @default.
- W3206136464 sameAs 3206136464 @default.
- W3206136464 citedByCount "3" @default.
- W3206136464 countsByYear W32061364642022 @default.
- W3206136464 countsByYear W32061364642023 @default.
- W3206136464 crossrefType "proceedings-article" @default.
- W3206136464 hasAuthorship W3206136464A5004837462 @default.
- W3206136464 hasAuthorship W3206136464A5035198517 @default.
- W3206136464 hasAuthorship W3206136464A5046414484 @default.
- W3206136464 hasAuthorship W3206136464A5067437853 @default.
- W3206136464 hasAuthorship W3206136464A5076699095 @default.
- W3206136464 hasAuthorship W3206136464A5076915400 @default.
- W3206136464 hasAuthorship W3206136464A5083700947 @default.
- W3206136464 hasConcept C101468663 @default.
- W3206136464 hasConcept C105795698 @default.
- W3206136464 hasConcept C121194460 @default.
- W3206136464 hasConcept C132525143 @default.
- W3206136464 hasConcept C149635348 @default.
- W3206136464 hasConcept C173608175 @default.
- W3206136464 hasConcept C18903297 @default.
- W3206136464 hasConcept C199360897 @default.
- W3206136464 hasConcept C2780165032 @default.
- W3206136464 hasConcept C33923547 @default.
- W3206136464 hasConcept C41008148 @default.
- W3206136464 hasConcept C42935608 @default.
- W3206136464 hasConcept C48044578 @default.
- W3206136464 hasConcept C68339613 @default.
- W3206136464 hasConcept C77088390 @default.
- W3206136464 hasConcept C80444323 @default.
- W3206136464 hasConcept C86803240 @default.
- W3206136464 hasConceptScore W3206136464C101468663 @default.
- W3206136464 hasConceptScore W3206136464C105795698 @default.
- W3206136464 hasConceptScore W3206136464C121194460 @default.
- W3206136464 hasConceptScore W3206136464C132525143 @default.
- W3206136464 hasConceptScore W3206136464C149635348 @default.
- W3206136464 hasConceptScore W3206136464C173608175 @default.
- W3206136464 hasConceptScore W3206136464C18903297 @default.
- W3206136464 hasConceptScore W3206136464C199360897 @default.
- W3206136464 hasConceptScore W3206136464C2780165032 @default.
- W3206136464 hasConceptScore W3206136464C33923547 @default.
- W3206136464 hasConceptScore W3206136464C41008148 @default.
- W3206136464 hasConceptScore W3206136464C42935608 @default.
- W3206136464 hasConceptScore W3206136464C48044578 @default.
- W3206136464 hasConceptScore W3206136464C68339613 @default.
- W3206136464 hasConceptScore W3206136464C77088390 @default.
- W3206136464 hasConceptScore W3206136464C80444323 @default.
- W3206136464 hasConceptScore W3206136464C86803240 @default.
- W3206136464 hasFunder F4320322927 @default.
- W3206136464 hasLocation W32061364641 @default.
- W3206136464 hasOpenAccess W3206136464 @default.
- W3206136464 hasPrimaryLocation W32061364641 @default.
- W3206136464 hasRelatedWork W157504995 @default.
- W3206136464 hasRelatedWork W1800827217 @default.
- W3206136464 hasRelatedWork W1967627035 @default.
- W3206136464 hasRelatedWork W1976363619 @default.
- W3206136464 hasRelatedWork W2095624907 @default.
- W3206136464 hasRelatedWork W2138261388 @default.
- W3206136464 hasRelatedWork W2164287667 @default.
- W3206136464 hasRelatedWork W2777249922 @default.
- W3206136464 hasRelatedWork W333278977 @default.
- W3206136464 hasRelatedWork W2078808231 @default.
- W3206136464 isParatext "false" @default.
- W3206136464 isRetracted "false" @default.
- W3206136464 magId "3206136464" @default.
- W3206136464 workType "article" @default.