Matches in SemOpenAlex for { <https://semopenalex.org/work/W2178700889> ?p ?o ?g. }
Showing items 1 to 60 of
60
with 100 items per page.
- W2178700889 abstract "Convolution operations have been widely used in many important application domains, such as deep learning and computer vision, in which convolution is always the most time-consuming part. High computational throughput and memory bandwidth make many-core architectures the promising targets to accelerate these applications. In this paper, we implement and optimize different convolution operations, including 1D convolution, 2D convolution and multi-channel 2D convolution executed in mini-batch mode, on both GPU and Intel MIC many-core architectures. We find out that the performance bottleneck of 1D and 2D convolutions is on registers rather than local memory or L1/L2 cache, and therefore, register tiling is used to improve the performance. In addition, we present a novel solution for multi-channel 2D convolution, in which convolution is conducted on images directly instead of being translated to matrix multiplication, and the data reuse of the algorithm is fully exploited. We further summarize the parameters of autotuning for multichannel 2D convolution and prune the search space based on heuristics. The experimental results show that, for the large filter size, our solution gets up to 33% performance improvement over cuDNN-v2 and up to 28% over clBLASbased implementation, on GTX TITAN and AMD W8000 respectively. On Intel MIC, our solution gets up to 25% of the theoretical peak performance." @default.
- W2178700889 created "2016-06-24" @default.
- W2178700889 creator A5001666028 @default.
- W2178700889 creator A5003680309 @default.
- W2178700889 creator A5024562387 @default.
- W2178700889 creator A5088172355 @default.
- W2178700889 date "2015-08-01" @default.
- W2178700889 modified "2023-09-24" @default.
- W2178700889 title "Fast Convolution Operations on Many-Core Architectures" @default.
- W2178700889 cites W1978642402 @default.
- W2178700889 cites W1984222112 @default.
- W2178700889 cites W2150606860 @default.
- W2178700889 cites W2155893237 @default.
- W2178700889 cites W2098547420 @default.
- W2178700889 doi "https://doi.org/10.1109/hpcc-css-icess.2015.94" @default.
- W2178700889 hasPublicationYear "2015" @default.
- W2178700889 type Work @default.
- W2178700889 sameAs 2178700889 @default.
- W2178700889 citedByCount "6" @default.
- W2178700889 countsByYear W21787008892017 @default.
- W2178700889 countsByYear W21787008892018 @default.
- W2178700889 countsByYear W21787008892019 @default.
- W2178700889 crossrefType "proceedings-article" @default.
- W2178700889 hasAuthorship W2178700889A5001666028 @default.
- W2178700889 hasAuthorship W2178700889A5003680309 @default.
- W2178700889 hasAuthorship W2178700889A5024562387 @default.
- W2178700889 hasAuthorship W2178700889A5088172355 @default.
- W2178700889 hasConcept C154945302 @default.
- W2178700889 hasConcept C173608175 @default.
- W2178700889 hasConcept C2164484 @default.
- W2178700889 hasConcept C41008148 @default.
- W2178700889 hasConcept C45347329 @default.
- W2178700889 hasConcept C50644808 @default.
- W2178700889 hasConcept C76155785 @default.
- W2178700889 hasConcept C79470037 @default.
- W2178700889 hasConceptScore W2178700889C154945302 @default.
- W2178700889 hasConceptScore W2178700889C173608175 @default.
- W2178700889 hasConceptScore W2178700889C2164484 @default.
- W2178700889 hasConceptScore W2178700889C41008148 @default.
- W2178700889 hasConceptScore W2178700889C45347329 @default.
- W2178700889 hasConceptScore W2178700889C50644808 @default.
- W2178700889 hasConceptScore W2178700889C76155785 @default.
- W2178700889 hasConceptScore W2178700889C79470037 @default.
- W2178700889 hasLocation W21787008891 @default.
- W2178700889 hasOpenAccess W2178700889 @default.
- W2178700889 hasPrimaryLocation W21787008891 @default.
- W2178700889 hasRelatedWork W1491899005 @default.
- W2178700889 hasRelatedWork W1502414128 @default.
- W2178700889 hasRelatedWork W1558545464 @default.
- W2178700889 hasRelatedWork W1604898313 @default.
- W2178700889 hasRelatedWork W2117014006 @default.
- W2178700889 hasRelatedWork W2172791042 @default.
- W2178700889 hasRelatedWork W2372170743 @default.
- W2178700889 hasRelatedWork W2391299576 @default.
- W2178700889 hasRelatedWork W2790489068 @default.
- W2178700889 hasRelatedWork W4233815414 @default.
- W2178700889 isParatext "false" @default.
- W2178700889 isRetracted "false" @default.
- W2178700889 magId "2178700889" @default.
- W2178700889 workType "article" @default.