Matches in SemOpenAlex for { <https://semopenalex.org/work/W4320062103> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4320062103 abstract "Integration of machine learning with simulation is part of a growing trend, however, the augmentation of codes in a highly-performant, distributed manner poses a software development challenge. In this work, we explore the question of how to easily augment legacy simulation codes on high-performance computers (HPCs) with machine-learned surrogate models, in a fast, scalable manner. Initial naïve augmentation attempts required significant code modification and resulted in significant slowdown. This led us to explore inference server techniques, which allow for model calls through drop-in functions. In this work, we investigated TensorFlow Serving with <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$mathbf{gRPC}$</tex> and RedisAI with SmartRedis for server-client inference implementations, where the deep learning platform runs as a persistent process on HPC compute node GPUs and the simulation makes client calls while running on the CPUs. We evaluated inference performance for several use cases on SCOUT, an IBM POWER9 supercomputer, including, real gas equations of state, machine-learned boundary conditions for rotorcraft aerodynamics, and super-resolution techniques. We will discuss key findings on performance. The lessons learned may provide useful advice for researchers to augment their simulation codes in an optimal manner." @default.
- W4320062103 created "2023-02-12" @default.
- W4320062103 creator A5006923065 @default.
- W4320062103 creator A5018102608 @default.
- W4320062103 creator A5032660760 @default.
- W4320062103 creator A5072676031 @default.
- W4320062103 date "2022-11-01" @default.
- W4320062103 modified "2023-10-03" @default.
- W4320062103 title "Scalable Integration of Computational Physics Simulations with Machine Learning" @default.
- W4320062103 cites W2030990315 @default.
- W4320062103 cites W2091423573 @default.
- W4320062103 cites W2133575275 @default.
- W4320062103 cites W2294710185 @default.
- W4320062103 cites W2937681869 @default.
- W4320062103 cites W3117082117 @default.
- W4320062103 cites W3134727219 @default.
- W4320062103 cites W3159903628 @default.
- W4320062103 cites W3168386838 @default.
- W4320062103 cites W4210921742 @default.
- W4320062103 cites W4285205277 @default.
- W4320062103 doi "https://doi.org/10.1109/ai4s56813.2022.00013" @default.
- W4320062103 hasPublicationYear "2022" @default.
- W4320062103 type Work @default.
- W4320062103 citedByCount "0" @default.
- W4320062103 crossrefType "proceedings-article" @default.
- W4320062103 hasAuthorship W4320062103A5006923065 @default.
- W4320062103 hasAuthorship W4320062103A5018102608 @default.
- W4320062103 hasAuthorship W4320062103A5032660760 @default.
- W4320062103 hasAuthorship W4320062103A5072676031 @default.
- W4320062103 hasConcept C111919701 @default.
- W4320062103 hasConcept C113775141 @default.
- W4320062103 hasConcept C118524514 @default.
- W4320062103 hasConcept C119857082 @default.
- W4320062103 hasConcept C154945302 @default.
- W4320062103 hasConcept C171250308 @default.
- W4320062103 hasConcept C173608175 @default.
- W4320062103 hasConcept C192562407 @default.
- W4320062103 hasConcept C199360897 @default.
- W4320062103 hasConcept C26517878 @default.
- W4320062103 hasConcept C26713055 @default.
- W4320062103 hasConcept C2776214188 @default.
- W4320062103 hasConcept C2777904410 @default.
- W4320062103 hasConcept C41008148 @default.
- W4320062103 hasConcept C48044578 @default.
- W4320062103 hasConcept C70388272 @default.
- W4320062103 hasConcept C83283714 @default.
- W4320062103 hasConcept C93996380 @default.
- W4320062103 hasConceptScore W4320062103C111919701 @default.
- W4320062103 hasConceptScore W4320062103C113775141 @default.
- W4320062103 hasConceptScore W4320062103C118524514 @default.
- W4320062103 hasConceptScore W4320062103C119857082 @default.
- W4320062103 hasConceptScore W4320062103C154945302 @default.
- W4320062103 hasConceptScore W4320062103C171250308 @default.
- W4320062103 hasConceptScore W4320062103C173608175 @default.
- W4320062103 hasConceptScore W4320062103C192562407 @default.
- W4320062103 hasConceptScore W4320062103C199360897 @default.
- W4320062103 hasConceptScore W4320062103C26517878 @default.
- W4320062103 hasConceptScore W4320062103C26713055 @default.
- W4320062103 hasConceptScore W4320062103C2776214188 @default.
- W4320062103 hasConceptScore W4320062103C2777904410 @default.
- W4320062103 hasConceptScore W4320062103C41008148 @default.
- W4320062103 hasConceptScore W4320062103C48044578 @default.
- W4320062103 hasConceptScore W4320062103C70388272 @default.
- W4320062103 hasConceptScore W4320062103C83283714 @default.
- W4320062103 hasConceptScore W4320062103C93996380 @default.
- W4320062103 hasLocation W43200621031 @default.
- W4320062103 hasOpenAccess W4320062103 @default.
- W4320062103 hasPrimaryLocation W43200621031 @default.
- W4320062103 hasRelatedWork W1487007137 @default.
- W4320062103 hasRelatedWork W1505993080 @default.
- W4320062103 hasRelatedWork W1878988619 @default.
- W4320062103 hasRelatedWork W2161718902 @default.
- W4320062103 hasRelatedWork W2470104108 @default.
- W4320062103 hasRelatedWork W2905887716 @default.
- W4320062103 hasRelatedWork W3120511008 @default.
- W4320062103 hasRelatedWork W3136744003 @default.
- W4320062103 hasRelatedWork W3195380914 @default.
- W4320062103 hasRelatedWork W3127959071 @default.
- W4320062103 isParatext "false" @default.
- W4320062103 isRetracted "false" @default.
- W4320062103 workType "article" @default.