FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1984506768> ?p ?o ?g. }

• W1984506768 endingPage "23" @default.
• W1984506768 startingPage "1" @default.
• W1984506768 abstract "The purpose of this research is to find a neural-network-based solution to the well-known problem of branch divergence in Single Instruction Multiple Data (SIMD) architectures. Our approach differs from existing techniques that handle branch (or control-flow) divergence, which use costly hardware modifications, low-utilization masking techniques, or static prediction methods. As we examine divergent applications, we characterize the degree of data-dependent control flow seen in each and isolate the code regions (or “kernels”) that cause the most performance degradation due to branch divergence. We then train neural networks (NNs) offline to approximate these kernels and inject the NN computations directly into the applications as substitutes for the kernels they approximate. This essentially translates control flow into nondivergent computation, trading off precision for performance. As our methodology manipulates application source code directly, it is inherently platform agnostic and can be adopted as a general means for accelerating divergent applications on data-parallel architectures. In this article, we present the Neuralizer, an automated software flow for kernel identification, NN training, and NN integration, as well as supplementary user-controlled optimization techniques. Evaluating our approach on a variety of divergent applications run on a Graphics Processing Unit (GPU), we on average achieve performance gains of 13.6 × and energy savings of 14.8 × with 96% accuracy." @default.
• W1984506768 created "2016-06-24" @default.
• W1984506768 creator A5059237344 @default.
• W1984506768 creator A5083493225 @default.
• W1984506768 date "2015-03-09" @default.
• W1984506768 modified "2023-10-16" @default.
• W1984506768 title "Accelerating Divergent Applications on SIMD Architectures Using Neural Networks" @default.
• W1984506768 cites W1498436455 @default.
• W1984506768 cites W1973538724 @default.
• W1984506768 cites W1981473264 @default.
• W1984506768 cites W1988115241 @default.
• W1984506768 cites W2010966003 @default.
• W1984506768 cites W2026764611 @default.
• W1984506768 cites W2037743346 @default.
• W1984506768 cites W2076304675 @default.
• W1984506768 cites W2089162427 @default.
• W1984506768 cites W2090584832 @default.
• W1984506768 cites W2105544671 @default.
• W1984506768 cites W2107220315 @default.
• W1984506768 cites W2114703523 @default.
• W1984506768 cites W2116267755 @default.
• W1984506768 cites W2119299853 @default.
• W1984506768 cites W2120585153 @default.
• W1984506768 cites W2128022558 @default.
• W1984506768 cites W2128317332 @default.
• W1984506768 cites W2132587889 @default.
• W1984506768 cites W2133218851 @default.
• W1984506768 cites W2135947393 @default.
• W1984506768 cites W2138565468 @default.
• W1984506768 cites W2142883190 @default.
• W1984506768 cites W2155503253 @default.
• W1984506768 cites W2156540297 @default.
• W1984506768 cites W2156831150 @default.
• W1984506768 cites W2157963512 @default.
• W1984506768 cites W2167399819 @default.
• W1984506768 cites W2169150396 @default.
• W1984506768 cites W2169875292 @default.
• W1984506768 cites W2169880332 @default.
• W1984506768 cites W2170881177 @default.
• W1984506768 cites W2187230075 @default.
• W1984506768 cites W4239437589 @default.
• W1984506768 cites W4240237526 @default.
• W1984506768 cites W4253998042 @default.
• W1984506768 cites W4361807832 @default.
• W1984506768 doi "https://doi.org/10.1145/2717311" @default.
• W1984506768 hasPublicationYear "2015" @default.
• W1984506768 type Work @default.
• W1984506768 sameAs 1984506768 @default.
• W1984506768 citedByCount "16" @default.
• W1984506768 countsByYear W19845067682016 @default.
• W1984506768 countsByYear W19845067682018 @default.
• W1984506768 countsByYear W19845067682019 @default.
• W1984506768 countsByYear W19845067682020 @default.
• W1984506768 countsByYear W19845067682021 @default.
• W1984506768 countsByYear W19845067682022 @default.
• W1984506768 crossrefType "journal-article" @default.
• W1984506768 hasAuthorship W1984506768A5059237344 @default.
• W1984506768 hasAuthorship W1984506768A5083493225 @default.
• W1984506768 hasBestOaLocation W19845067681 @default.
• W1984506768 hasConcept C113775141 @default.
• W1984506768 hasConcept C11413529 @default.
• W1984506768 hasConcept C114614502 @default.
• W1984506768 hasConcept C119857082 @default.
• W1984506768 hasConcept C138885662 @default.
• W1984506768 hasConcept C139571649 @default.
• W1984506768 hasConcept C150552126 @default.
• W1984506768 hasConcept C160191386 @default.
• W1984506768 hasConcept C169590947 @default.
• W1984506768 hasConcept C173608175 @default.
• W1984506768 hasConcept C177264268 @default.
• W1984506768 hasConcept C199360897 @default.
• W1984506768 hasConcept C207390915 @default.
• W1984506768 hasConcept C2776760102 @default.
• W1984506768 hasConcept C2777904410 @default.
• W1984506768 hasConcept C2779851693 @default.
• W1984506768 hasConcept C33923547 @default.
• W1984506768 hasConcept C41008148 @default.
• W1984506768 hasConcept C41895202 @default.
• W1984506768 hasConcept C43126263 @default.
• W1984506768 hasConcept C45374587 @default.
• W1984506768 hasConcept C50644808 @default.
• W1984506768 hasConcept C74193536 @default.
• W1984506768 hasConceptScore W1984506768C113775141 @default.
• W1984506768 hasConceptScore W1984506768C11413529 @default.
• W1984506768 hasConceptScore W1984506768C114614502 @default.
• W1984506768 hasConceptScore W1984506768C119857082 @default.
• W1984506768 hasConceptScore W1984506768C138885662 @default.
• W1984506768 hasConceptScore W1984506768C139571649 @default.
• W1984506768 hasConceptScore W1984506768C150552126 @default.
• W1984506768 hasConceptScore W1984506768C160191386 @default.
• W1984506768 hasConceptScore W1984506768C169590947 @default.
• W1984506768 hasConceptScore W1984506768C173608175 @default.
• W1984506768 hasConceptScore W1984506768C177264268 @default.
• W1984506768 hasConceptScore W1984506768C199360897 @default.
• W1984506768 hasConceptScore W1984506768C207390915 @default.
• W1984506768 hasConceptScore W1984506768C2776760102 @default.
• W1984506768 hasConceptScore W1984506768C2777904410 @default.
• W1984506768 hasConceptScore W1984506768C2779851693 @default.

• next