Matches in SemOpenAlex for { <https://semopenalex.org/work/W4320062111> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W4320062111 abstract "Stagnation of Moore's law has led to the increased adoption of parallel programming for enhancing performance of scientific applications. Frequently occurring code and design patterns in scientific applications are often used for transforming serial code to parallel. But, identifying these patterns is not easy. To this end, we propose using Graph Neural Networks for modeling code flow graphs to identify patterns in such parallel code. Additionally, identifying the runtime parameters for best performing parallel code is also challenging. We propose a pattern-guided deep learning based tuning approach, to help identify the best runtime parameters for OpenMP loops. Overall, we aim to identify commonly occurring patterns in parallel loops and use these patterns to guide auto-tuning efforts. We validate our hypothesis on 20 different applications from Polybench, and STREAM benchmark suites. This deep learning-based approach can identify the considered patterns with an overall accuracy of 91%. We validate the usefulness of using patterns for auto-tuning on tuning the number of threads, scheduling policies and chunk size on a single socket system, and the thread count and affinity on a multi-socket machine. Our approach achieves geometric mean speedups of <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$1.1times$</tex> and <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$4.7times$</tex> respectively over default OpenMP configurations, compared to brute-force speedups of <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$1.27times$</tex> and <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$4.93times$</tex> respectively." @default.
- W4320062111 created "2023-02-12" @default.
- W4320062111 creator A5026497457 @default.
- W4320062111 creator A5027335690 @default.
- W4320062111 creator A5063385499 @default.
- W4320062111 creator A5064997043 @default.
- W4320062111 creator A5079359777 @default.
- W4320062111 creator A5090330391 @default.
- W4320062111 date "2022-11-01" @default.
- W4320062111 modified "2023-09-27" @default.
- W4320062111 title "Pattern-based Autotuning of OpenMP Loops using Graph Neural Networks" @default.
- W4320062111 cites W1979841749 @default.
- W4320062111 cites W1987564528 @default.
- W4320062111 cites W2034761517 @default.
- W4320062111 cites W2140131713 @default.
- W4320062111 cites W2175209623 @default.
- W4320062111 cites W2483923764 @default.
- W4320062111 cites W2604314403 @default.
- W4320062111 cites W2963935794 @default.
- W4320062111 cites W2967037890 @default.
- W4320062111 cites W2969583281 @default.
- W4320062111 cites W2995287845 @default.
- W4320062111 cites W3005951744 @default.
- W4320062111 cites W3007205676 @default.
- W4320062111 cites W3116350821 @default.
- W4320062111 cites W3131559736 @default.
- W4320062111 cites W3157407232 @default.
- W4320062111 cites W3168986600 @default.
- W4320062111 cites W4210257598 @default.
- W4320062111 cites W4225275160 @default.
- W4320062111 cites W4285503927 @default.
- W4320062111 doi "https://doi.org/10.1109/ai4s56813.2022.00010" @default.
- W4320062111 hasPublicationYear "2022" @default.
- W4320062111 type Work @default.
- W4320062111 citedByCount "3" @default.
- W4320062111 countsByYear W43200621112023 @default.
- W4320062111 crossrefType "proceedings-article" @default.
- W4320062111 hasAuthorship W4320062111A5026497457 @default.
- W4320062111 hasAuthorship W4320062111A5027335690 @default.
- W4320062111 hasAuthorship W4320062111A5063385499 @default.
- W4320062111 hasAuthorship W4320062111A5064997043 @default.
- W4320062111 hasAuthorship W4320062111A5079359777 @default.
- W4320062111 hasAuthorship W4320062111A5090330391 @default.
- W4320062111 hasConcept C13280743 @default.
- W4320062111 hasConcept C138101251 @default.
- W4320062111 hasConcept C154945302 @default.
- W4320062111 hasConcept C162324750 @default.
- W4320062111 hasConcept C173608175 @default.
- W4320062111 hasConcept C177264268 @default.
- W4320062111 hasConcept C185798385 @default.
- W4320062111 hasConcept C199360897 @default.
- W4320062111 hasConcept C205649164 @default.
- W4320062111 hasConcept C206729178 @default.
- W4320062111 hasConcept C21547014 @default.
- W4320062111 hasConcept C2776760102 @default.
- W4320062111 hasConcept C41008148 @default.
- W4320062111 hasConcept C50644808 @default.
- W4320062111 hasConceptScore W4320062111C13280743 @default.
- W4320062111 hasConceptScore W4320062111C138101251 @default.
- W4320062111 hasConceptScore W4320062111C154945302 @default.
- W4320062111 hasConceptScore W4320062111C162324750 @default.
- W4320062111 hasConceptScore W4320062111C173608175 @default.
- W4320062111 hasConceptScore W4320062111C177264268 @default.
- W4320062111 hasConceptScore W4320062111C185798385 @default.
- W4320062111 hasConceptScore W4320062111C199360897 @default.
- W4320062111 hasConceptScore W4320062111C205649164 @default.
- W4320062111 hasConceptScore W4320062111C206729178 @default.
- W4320062111 hasConceptScore W4320062111C21547014 @default.
- W4320062111 hasConceptScore W4320062111C2776760102 @default.
- W4320062111 hasConceptScore W4320062111C41008148 @default.
- W4320062111 hasConceptScore W4320062111C50644808 @default.
- W4320062111 hasFunder F4320321505 @default.
- W4320062111 hasFunder F4320322930 @default.
- W4320062111 hasLocation W43200621111 @default.
- W4320062111 hasOpenAccess W4320062111 @default.
- W4320062111 hasPrimaryLocation W43200621111 @default.
- W4320062111 hasRelatedWork W112744582 @default.
- W4320062111 hasRelatedWork W1485630101 @default.
- W4320062111 hasRelatedWork W1579918296 @default.
- W4320062111 hasRelatedWork W2133693067 @default.
- W4320062111 hasRelatedWork W2170268965 @default.
- W4320062111 hasRelatedWork W2259585419 @default.
- W4320062111 hasRelatedWork W2498017833 @default.
- W4320062111 hasRelatedWork W2893160992 @default.
- W4320062111 hasRelatedWork W3033052750 @default.
- W4320062111 hasRelatedWork W4210381016 @default.
- W4320062111 isParatext "false" @default.
- W4320062111 isRetracted "false" @default.
- W4320062111 workType "article" @default.