Matches in SemOpenAlex for { <https://semopenalex.org/work/W4315706831> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W4315706831 abstract "Building large AI fleets to support the rapidly growing DL workloads is an active research topic for modern cloud providers. Generating accurate benchmarks plays an essential role in designing the fast-paced software and hardware solutions in this space. Two fundamental challenges to make this scalable are (i) workload representativeness and (ii) the ability to quickly incorporate changes to the fleet into the benchmarks. To overcome these issues, we propose Mystique, an accurate and scalable framework for production AI benchmark generation. It leverages the PyTorch execution trace (ET), a new feature that captures the runtime information of AI models at the granularity of operators, in a graph format, together with their metadata. By sourcing fleet ETs, we can build AI benchmarks that are portable and representative. Mystique is scalable, due to its lightweight data collection, in terms of runtime overhead and instrumentation effort. It is also adaptive because ET composability allows flexible control on benchmark creation. We evaluate our methodology on several production AI models, and show that benchmarks generated with Mystique closely resemble original AI models, both in execution time and system-level metrics. We also showcase the portability of the generated benchmarks across platforms, and demonstrate several use cases enabled by the fine-grained composability of the execution trace." @default.
- W4315706831 created "2023-01-12" @default.
- W4315706831 creator A5001498296 @default.
- W4315706831 creator A5008646988 @default.
- W4315706831 creator A5019271298 @default.
- W4315706831 creator A5034003919 @default.
- W4315706831 creator A5039120403 @default.
- W4315706831 creator A5081888342 @default.
- W4315706831 creator A5087652753 @default.
- W4315706831 date "2022-12-16" @default.
- W4315706831 modified "2023-09-27" @default.
- W4315706831 title "Mystique: Enabling Accurate and Scalable Generation of Production AI Benchmarks" @default.
- W4315706831 doi "https://doi.org/10.48550/arxiv.2301.04122" @default.
- W4315706831 hasPublicationYear "2022" @default.
- W4315706831 type Work @default.
- W4315706831 citedByCount "0" @default.
- W4315706831 crossrefType "posted-content" @default.
- W4315706831 hasAuthorship W4315706831A5001498296 @default.
- W4315706831 hasAuthorship W4315706831A5008646988 @default.
- W4315706831 hasAuthorship W4315706831A5019271298 @default.
- W4315706831 hasAuthorship W4315706831A5034003919 @default.
- W4315706831 hasAuthorship W4315706831A5039120403 @default.
- W4315706831 hasAuthorship W4315706831A5081888342 @default.
- W4315706831 hasAuthorship W4315706831A5087652753 @default.
- W4315706831 hasBestOaLocation W43157068311 @default.
- W4315706831 hasConcept C111919701 @default.
- W4315706831 hasConcept C118524514 @default.
- W4315706831 hasConcept C120314980 @default.
- W4315706831 hasConcept C13280743 @default.
- W4315706831 hasConcept C138885662 @default.
- W4315706831 hasConcept C169590947 @default.
- W4315706831 hasConcept C177774035 @default.
- W4315706831 hasConcept C185798385 @default.
- W4315706831 hasConcept C199360897 @default.
- W4315706831 hasConcept C205649164 @default.
- W4315706831 hasConcept C2777904410 @default.
- W4315706831 hasConcept C2778814252 @default.
- W4315706831 hasConcept C2779960059 @default.
- W4315706831 hasConcept C31258907 @default.
- W4315706831 hasConcept C31395832 @default.
- W4315706831 hasConcept C41008148 @default.
- W4315706831 hasConcept C41895202 @default.
- W4315706831 hasConcept C48044578 @default.
- W4315706831 hasConcept C63000827 @default.
- W4315706831 hasConcept C75291252 @default.
- W4315706831 hasConcept C77088390 @default.
- W4315706831 hasConcept C79974875 @default.
- W4315706831 hasConcept C93518851 @default.
- W4315706831 hasConceptScore W4315706831C111919701 @default.
- W4315706831 hasConceptScore W4315706831C118524514 @default.
- W4315706831 hasConceptScore W4315706831C120314980 @default.
- W4315706831 hasConceptScore W4315706831C13280743 @default.
- W4315706831 hasConceptScore W4315706831C138885662 @default.
- W4315706831 hasConceptScore W4315706831C169590947 @default.
- W4315706831 hasConceptScore W4315706831C177774035 @default.
- W4315706831 hasConceptScore W4315706831C185798385 @default.
- W4315706831 hasConceptScore W4315706831C199360897 @default.
- W4315706831 hasConceptScore W4315706831C205649164 @default.
- W4315706831 hasConceptScore W4315706831C2777904410 @default.
- W4315706831 hasConceptScore W4315706831C2778814252 @default.
- W4315706831 hasConceptScore W4315706831C2779960059 @default.
- W4315706831 hasConceptScore W4315706831C31258907 @default.
- W4315706831 hasConceptScore W4315706831C31395832 @default.
- W4315706831 hasConceptScore W4315706831C41008148 @default.
- W4315706831 hasConceptScore W4315706831C41895202 @default.
- W4315706831 hasConceptScore W4315706831C48044578 @default.
- W4315706831 hasConceptScore W4315706831C63000827 @default.
- W4315706831 hasConceptScore W4315706831C75291252 @default.
- W4315706831 hasConceptScore W4315706831C77088390 @default.
- W4315706831 hasConceptScore W4315706831C79974875 @default.
- W4315706831 hasConceptScore W4315706831C93518851 @default.
- W4315706831 hasLocation W43157068311 @default.
- W4315706831 hasOpenAccess W4315706831 @default.
- W4315706831 hasPrimaryLocation W43157068311 @default.
- W4315706831 hasRelatedWork W1911735095 @default.
- W4315706831 hasRelatedWork W1987753593 @default.
- W4315706831 hasRelatedWork W1992807924 @default.
- W4315706831 hasRelatedWork W1995399085 @default.
- W4315706831 hasRelatedWork W2267601986 @default.
- W4315706831 hasRelatedWork W2380023786 @default.
- W4315706831 hasRelatedWork W2475198316 @default.
- W4315706831 hasRelatedWork W2586664332 @default.
- W4315706831 hasRelatedWork W4300361016 @default.
- W4315706831 hasRelatedWork W4364383452 @default.
- W4315706831 isParatext "false" @default.
- W4315706831 isRetracted "false" @default.
- W4315706831 workType "article" @default.