Matches in SemOpenAlex for { <https://semopenalex.org/work/W3126298766> ?p ?o ?g. }
- W3126298766 abstract "Nowadays, GPU accelerators are commonly used to speed up general-purpose computing tasks on a variety of hardware. However, due to the diversity of GPU architectures and processed data, optimization of codes for a particular type of hardware and specific data characteristics can be extremely challenging. The autotuning of performance-relevant source-code parameters allows for automatic optimization of applications and keeps their performance portable. Although the autotuning process typically results in code speed-up, searching the tuning space can bring unacceptable overhead if (i) the tuning space is vast and full of poorly-performing implementations, or (ii) the autotuning process has to be repeated frequently because of changes in processed data or migration to different hardware. In this paper, we introduce a novel method for searching tuning spaces. The method takes advantage of collecting hardware performance counters (also known as profiling counters) during empirical tuning. Those counters are used to navigate the searching process towards faster implementations. The method requires the tuning space to be sampled on any GPU. It builds a problem-specific model, which can be used during autotuning on various, even previously unseen inputs or GPUs. Using a set of five benchmarks, we experimentally demonstrate that our method can speed up autotuning when an application needs to be ported to different hardware or when it needs to process data with different characteristics. We also compared our method to state of the art and show that our method is superior in terms of the number of searching steps and typically outperforms other searches in terms of convergence time." @default.
- W3126298766 created "2021-02-15" @default.
- W3126298766 creator A5005015118 @default.
- W3126298766 creator A5054834154 @default.
- W3126298766 creator A5067483779 @default.
- W3126298766 creator A5075255082 @default.
- W3126298766 creator A5089216324 @default.
- W3126298766 date "2021-02-10" @default.
- W3126298766 modified "2023-09-27" @default.
- W3126298766 title "Using hardware performance counters to speed up autotuning convergence on GPUs" @default.
- W3126298766 cites W1967701350 @default.
- W3126298766 cites W1978642402 @default.
- W3126298766 cites W1979527452 @default.
- W3126298766 cites W2036856982 @default.
- W3126298766 cites W2038666141 @default.
- W3126298766 cites W2045128810 @default.
- W3126298766 cites W2063750261 @default.
- W3126298766 cites W2070544163 @default.
- W3126298766 cites W2100218206 @default.
- W3126298766 cites W2113282196 @default.
- W3126298766 cites W2130336316 @default.
- W3126298766 cites W2136440628 @default.
- W3126298766 cites W2142079700 @default.
- W3126298766 cites W2142769604 @default.
- W3126298766 cites W2144264070 @default.
- W3126298766 cites W2149706766 @default.
- W3126298766 cites W2167334577 @default.
- W3126298766 cites W2314321304 @default.
- W3126298766 cites W2329047703 @default.
- W3126298766 cites W2480834041 @default.
- W3126298766 cites W2534888058 @default.
- W3126298766 cites W2730174870 @default.
- W3126298766 cites W2768520977 @default.
- W3126298766 cites W2784409032 @default.
- W3126298766 cites W2785611382 @default.
- W3126298766 cites W2786865931 @default.
- W3126298766 cites W2887327791 @default.
- W3126298766 cites W2967037890 @default.
- W3126298766 cites W2971086374 @default.
- W3126298766 cites W3048408968 @default.
- W3126298766 cites W3102753670 @default.
- W3126298766 cites W3215765506 @default.
- W3126298766 cites W3010412873 @default.
- W3126298766 hasPublicationYear "2021" @default.
- W3126298766 type Work @default.
- W3126298766 sameAs 3126298766 @default.
- W3126298766 citedByCount "0" @default.
- W3126298766 crossrefType "posted-content" @default.
- W3126298766 hasAuthorship W3126298766A5005015118 @default.
- W3126298766 hasAuthorship W3126298766A5054834154 @default.
- W3126298766 hasAuthorship W3126298766A5067483779 @default.
- W3126298766 hasAuthorship W3126298766A5075255082 @default.
- W3126298766 hasAuthorship W3126298766A5089216324 @default.
- W3126298766 hasConcept C106251023 @default.
- W3126298766 hasConcept C113775141 @default.
- W3126298766 hasConcept C173608175 @default.
- W3126298766 hasConcept C177264268 @default.
- W3126298766 hasConcept C187191949 @default.
- W3126298766 hasConcept C199360897 @default.
- W3126298766 hasConcept C26713055 @default.
- W3126298766 hasConcept C2776760102 @default.
- W3126298766 hasConcept C2777904410 @default.
- W3126298766 hasConcept C2779960059 @default.
- W3126298766 hasConcept C41008148 @default.
- W3126298766 hasConcept C68339613 @default.
- W3126298766 hasConcept C9390403 @default.
- W3126298766 hasConcept C98045186 @default.
- W3126298766 hasConceptScore W3126298766C106251023 @default.
- W3126298766 hasConceptScore W3126298766C113775141 @default.
- W3126298766 hasConceptScore W3126298766C173608175 @default.
- W3126298766 hasConceptScore W3126298766C177264268 @default.
- W3126298766 hasConceptScore W3126298766C187191949 @default.
- W3126298766 hasConceptScore W3126298766C199360897 @default.
- W3126298766 hasConceptScore W3126298766C26713055 @default.
- W3126298766 hasConceptScore W3126298766C2776760102 @default.
- W3126298766 hasConceptScore W3126298766C2777904410 @default.
- W3126298766 hasConceptScore W3126298766C2779960059 @default.
- W3126298766 hasConceptScore W3126298766C41008148 @default.
- W3126298766 hasConceptScore W3126298766C68339613 @default.
- W3126298766 hasConceptScore W3126298766C9390403 @default.
- W3126298766 hasConceptScore W3126298766C98045186 @default.
- W3126298766 hasLocation W31262987661 @default.
- W3126298766 hasOpenAccess W3126298766 @default.
- W3126298766 hasPrimaryLocation W31262987661 @default.
- W3126298766 hasRelatedWork W1795216021 @default.
- W3126298766 hasRelatedWork W1984382863 @default.
- W3126298766 hasRelatedWork W2009451155 @default.
- W3126298766 hasRelatedWork W2172971060 @default.
- W3126298766 hasRelatedWork W2344642201 @default.
- W3126298766 hasRelatedWork W2563030477 @default.
- W3126298766 hasRelatedWork W2671496873 @default.
- W3126298766 hasRelatedWork W2732023272 @default.
- W3126298766 hasRelatedWork W2754990314 @default.
- W3126298766 hasRelatedWork W2783431496 @default.
- W3126298766 hasRelatedWork W2785611382 @default.
- W3126298766 hasRelatedWork W2922025187 @default.
- W3126298766 hasRelatedWork W2948530470 @default.
- W3126298766 hasRelatedWork W2951815798 @default.
- W3126298766 hasRelatedWork W2953190515 @default.
- W3126298766 hasRelatedWork W2969451848 @default.