Matches in SemOpenAlex for { <https://semopenalex.org/work/W2756531049> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W2756531049 abstract "A heterogeneous memory system (HMS) consists of multiple memory components with different properties. GPU is a representative architecture with HMS. It is challenging to decide optimal placement of data objects on HMS because of the large exploration space and complicated memory hierarchy on HMS. In this paper, we introduce performance modeling techniques to predict performance of various data placements on GPU. In essence, our models quantify and capture implicit performance correlation between different data placements. Given the memory access information and performance of a sample data placement, our models predict performance for other data placements based on the quantified correlation. We reveal critical performance factors that cause performance variation across data placements. Those factors include instruction replay, addressing mode, hardware queuing delay of memory requests, off-chip memory access latency, and caching effects. Those factors, which are often not sufficiently considered in the existing performance models, can significantly impact modeling accuracy. We introduce a series of techniques to model those factors. We extensively evaluate our models with a variety of benchmarks with various data placements. Our models are able to quantitatively predict the benefit or performance loss of data placements." @default.
- W2756531049 created "2017-10-06" @default.
- W2756531049 creator A5056036725 @default.
- W2756531049 creator A5056846237 @default.
- W2756531049 date "2017-09-01" @default.
- W2756531049 modified "2023-09-26" @default.
- W2756531049 title "Performance Modeling for Optimal Data Placement on GPU with Heterogeneous Memory Systems" @default.
- W2756531049 cites W1966774236 @default.
- W2756531049 cites W1979527452 @default.
- W2756531049 cites W1992851788 @default.
- W2756531049 cites W2033486618 @default.
- W2756531049 cites W2038101378 @default.
- W2756531049 cites W2091553595 @default.
- W2756531049 cites W2098290747 @default.
- W2756531049 cites W2101320423 @default.
- W2756531049 cites W2102843684 @default.
- W2756531049 cites W2111045724 @default.
- W2756531049 cites W2113282196 @default.
- W2756531049 cites W2115172404 @default.
- W2756531049 cites W2120062497 @default.
- W2756531049 cites W2124237318 @default.
- W2756531049 cites W2128539477 @default.
- W2756531049 cites W2130336316 @default.
- W2756531049 cites W2142769604 @default.
- W2756531049 cites W2144264070 @default.
- W2756531049 cites W2147193503 @default.
- W2756531049 cites W2149234156 @default.
- W2756531049 cites W2167334577 @default.
- W2756531049 cites W2316410352 @default.
- W2756531049 cites W2412449650 @default.
- W2756531049 cites W4246740707 @default.
- W2756531049 cites W4255416366 @default.
- W2756531049 doi "https://doi.org/10.1109/cluster.2017.42" @default.
- W2756531049 hasPublicationYear "2017" @default.
- W2756531049 type Work @default.
- W2756531049 sameAs 2756531049 @default.
- W2756531049 citedByCount "9" @default.
- W2756531049 countsByYear W27565310492018 @default.
- W2756531049 countsByYear W27565310492019 @default.
- W2756531049 countsByYear W27565310492020 @default.
- W2756531049 countsByYear W27565310492022 @default.
- W2756531049 crossrefType "proceedings-article" @default.
- W2756531049 hasAuthorship W2756531049A5056036725 @default.
- W2756531049 hasAuthorship W2756531049A5056846237 @default.
- W2756531049 hasConcept C111919701 @default.
- W2756531049 hasConcept C115537543 @default.
- W2756531049 hasConcept C118524514 @default.
- W2756531049 hasConcept C120314980 @default.
- W2756531049 hasConcept C162319229 @default.
- W2756531049 hasConcept C173608175 @default.
- W2756531049 hasConcept C199360897 @default.
- W2756531049 hasConcept C22684755 @default.
- W2756531049 hasConcept C2777115002 @default.
- W2756531049 hasConcept C2778100165 @default.
- W2756531049 hasConcept C31258907 @default.
- W2756531049 hasConcept C41008148 @default.
- W2756531049 hasConcept C47487241 @default.
- W2756531049 hasConcept C76155785 @default.
- W2756531049 hasConcept C77088390 @default.
- W2756531049 hasConcept C82876162 @default.
- W2756531049 hasConceptScore W2756531049C111919701 @default.
- W2756531049 hasConceptScore W2756531049C115537543 @default.
- W2756531049 hasConceptScore W2756531049C118524514 @default.
- W2756531049 hasConceptScore W2756531049C120314980 @default.
- W2756531049 hasConceptScore W2756531049C162319229 @default.
- W2756531049 hasConceptScore W2756531049C173608175 @default.
- W2756531049 hasConceptScore W2756531049C199360897 @default.
- W2756531049 hasConceptScore W2756531049C22684755 @default.
- W2756531049 hasConceptScore W2756531049C2777115002 @default.
- W2756531049 hasConceptScore W2756531049C2778100165 @default.
- W2756531049 hasConceptScore W2756531049C31258907 @default.
- W2756531049 hasConceptScore W2756531049C41008148 @default.
- W2756531049 hasConceptScore W2756531049C47487241 @default.
- W2756531049 hasConceptScore W2756531049C76155785 @default.
- W2756531049 hasConceptScore W2756531049C77088390 @default.
- W2756531049 hasConceptScore W2756531049C82876162 @default.
- W2756531049 hasLocation W27565310491 @default.
- W2756531049 hasOpenAccess W2756531049 @default.
- W2756531049 hasPrimaryLocation W27565310491 @default.
- W2756531049 hasRelatedWork W1509211761 @default.
- W2756531049 hasRelatedWork W1566886392 @default.
- W2756531049 hasRelatedWork W2122052575 @default.
- W2756531049 hasRelatedWork W2298102683 @default.
- W2756531049 hasRelatedWork W2383341208 @default.
- W2756531049 hasRelatedWork W2588901507 @default.
- W2756531049 hasRelatedWork W2756531049 @default.
- W2756531049 hasRelatedWork W3033499831 @default.
- W2756531049 hasRelatedWork W4232453487 @default.
- W2756531049 hasRelatedWork W4245302940 @default.
- W2756531049 isParatext "false" @default.
- W2756531049 isRetracted "false" @default.
- W2756531049 magId "2756531049" @default.
- W2756531049 workType "article" @default.