Matches in SemOpenAlex for { <https://semopenalex.org/work/W3034870003> ?p ?o ?g. }
- W3034870003 endingPage "1857" @default.
- W3034870003 startingPage "1844" @default.
- W3034870003 abstract "Data processing on convolutional neural networks (CNNs) places a heavy burden on energy-constrained mobile platforms. This article optimizes energy on a mobile client by partitioning CNN computations between in situ processing on the client and offloaded computations in the cloud. A new analytical CNN energy model is formulated, capturing all major components of the in situ computation, for ASIC-based deep learning accelerators. The model is benchmarked against measured silicon data. The analytical framework is used to determine the optimal energy partition point between the client and the cloud at runtime. On standard CNN topologies, partitioned computation is demonstrated to provide significant energy savings on the client over a fully cloud-based computation or fully in situ computation. For example, at 80 Mbps effective bit rate and 0.78 W transmission power, the optimal partition for AlexNet [SqueezeNet] saves up to 52.4% [73.4%] energy over a fully cloud-based computation and 27.3% [28.8%] energy over a fully in situ computation." @default.
- W3034870003 created "2020-06-19" @default.
- W3034870003 creator A5004590102 @default.
- W3034870003 creator A5087094576 @default.
- W3034870003 creator A5090685279 @default.
- W3034870003 date "2020-08-01" @default.
- W3034870003 modified "2023-10-17" @default.
- W3034870003 title "NeuPart: Using Analytical Models to Drive Energy-Efficient Partitioning of CNN Computations on Cloud-Connected Mobile Clients" @default.
- W3034870003 cites W1801710398 @default.
- W3034870003 cites W1976905670 @default.
- W3034870003 cites W2012423440 @default.
- W3034870003 cites W2067523571 @default.
- W3034870003 cites W2097117768 @default.
- W3034870003 cites W2128404326 @default.
- W3034870003 cites W2144651529 @default.
- W3034870003 cites W2150073849 @default.
- W3034870003 cites W2289252105 @default.
- W3034870003 cites W2290132443 @default.
- W3034870003 cites W2402093298 @default.
- W3034870003 cites W2442974303 @default.
- W3034870003 cites W2473156356 @default.
- W3034870003 cites W2581082771 @default.
- W3034870003 cites W2604319603 @default.
- W3034870003 cites W2605258629 @default.
- W3034870003 cites W2606722458 @default.
- W3034870003 cites W2623629680 @default.
- W3034870003 cites W2786070938 @default.
- W3034870003 cites W2800920215 @default.
- W3034870003 cites W2806506455 @default.
- W3034870003 cites W2892952080 @default.
- W3034870003 cites W2920031528 @default.
- W3034870003 cites W2929565236 @default.
- W3034870003 cites W2949628230 @default.
- W3034870003 cites W2963367920 @default.
- W3034870003 cites W4236302577 @default.
- W3034870003 doi "https://doi.org/10.1109/tvlsi.2020.2995135" @default.
- W3034870003 hasPublicationYear "2020" @default.
- W3034870003 type Work @default.
- W3034870003 sameAs 3034870003 @default.
- W3034870003 citedByCount "6" @default.
- W3034870003 countsByYear W30348700032021 @default.
- W3034870003 countsByYear W30348700032022 @default.
- W3034870003 countsByYear W30348700032023 @default.
- W3034870003 crossrefType "journal-article" @default.
- W3034870003 hasAuthorship W3034870003A5004590102 @default.
- W3034870003 hasAuthorship W3034870003A5087094576 @default.
- W3034870003 hasAuthorship W3034870003A5090685279 @default.
- W3034870003 hasBestOaLocation W30348700031 @default.
- W3034870003 hasConcept C111919701 @default.
- W3034870003 hasConcept C11413529 @default.
- W3034870003 hasConcept C114614502 @default.
- W3034870003 hasConcept C119599485 @default.
- W3034870003 hasConcept C127413603 @default.
- W3034870003 hasConcept C149635348 @default.
- W3034870003 hasConcept C154945302 @default.
- W3034870003 hasConcept C173608175 @default.
- W3034870003 hasConcept C186967261 @default.
- W3034870003 hasConcept C2742236 @default.
- W3034870003 hasConcept C2778456923 @default.
- W3034870003 hasConcept C2779191767 @default.
- W3034870003 hasConcept C2780165032 @default.
- W3034870003 hasConcept C2781041963 @default.
- W3034870003 hasConcept C33923547 @default.
- W3034870003 hasConcept C41008148 @default.
- W3034870003 hasConcept C42812 @default.
- W3034870003 hasConcept C45374587 @default.
- W3034870003 hasConcept C459310 @default.
- W3034870003 hasConcept C77390884 @default.
- W3034870003 hasConcept C79403827 @default.
- W3034870003 hasConcept C79974875 @default.
- W3034870003 hasConcept C81363708 @default.
- W3034870003 hasConceptScore W3034870003C111919701 @default.
- W3034870003 hasConceptScore W3034870003C11413529 @default.
- W3034870003 hasConceptScore W3034870003C114614502 @default.
- W3034870003 hasConceptScore W3034870003C119599485 @default.
- W3034870003 hasConceptScore W3034870003C127413603 @default.
- W3034870003 hasConceptScore W3034870003C149635348 @default.
- W3034870003 hasConceptScore W3034870003C154945302 @default.
- W3034870003 hasConceptScore W3034870003C173608175 @default.
- W3034870003 hasConceptScore W3034870003C186967261 @default.
- W3034870003 hasConceptScore W3034870003C2742236 @default.
- W3034870003 hasConceptScore W3034870003C2778456923 @default.
- W3034870003 hasConceptScore W3034870003C2779191767 @default.
- W3034870003 hasConceptScore W3034870003C2780165032 @default.
- W3034870003 hasConceptScore W3034870003C2781041963 @default.
- W3034870003 hasConceptScore W3034870003C33923547 @default.
- W3034870003 hasConceptScore W3034870003C41008148 @default.
- W3034870003 hasConceptScore W3034870003C42812 @default.
- W3034870003 hasConceptScore W3034870003C45374587 @default.
- W3034870003 hasConceptScore W3034870003C459310 @default.
- W3034870003 hasConceptScore W3034870003C77390884 @default.
- W3034870003 hasConceptScore W3034870003C79403827 @default.
- W3034870003 hasConceptScore W3034870003C79974875 @default.
- W3034870003 hasConceptScore W3034870003C81363708 @default.
- W3034870003 hasFunder F4320306076 @default.
- W3034870003 hasIssue "8" @default.
- W3034870003 hasLocation W30348700031 @default.
- W3034870003 hasLocation W30348700032 @default.