FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3085005384> ?p ?o ?g. }

• W3085005384 endingPage "1" @default.
• W3085005384 startingPage "1" @default.
• W3085005384 abstract "The performance of a task running on a many-core with distributed shared last-level cache (LLC) strongly depends on two parameters: the power budget needed to guarantee thermally-safe operation and the LLC latency. The task's thread-to-core mapping determines both the parameters and needs to make a trade-off because both cannot be simultaneously optimal. Arrival and departure of tasks on a many-core deployed in an open system can change its state significantly in terms of available cores and power budgets. Task migrations can thereupon be used as a tool to keep the many-core operating at peak performance. Furthermore, the relative impacts of power budget and LLC latency on a task's performance may change with its different execution phases mandating its migration on-the-fly. We propose the first run-time algorithm <italic xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>PCMig</i> that increases the performance of a many-core with distributed shared LLC by migrating tasks based on their phases and the many-core's state. <italic xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>PCMig</i> is based on a model that predicts the performance impact of migrations. We propose a performance prediction model based on a lightweight neural network (NN). To serve as a reference, we also propose an analytical model of the many-core that operates on CPI stacks. We demonstrate an NN-based model achieves a higher prediction accuracy at a lower overhead than an analytical model. <italic xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>PCMig</i> is based on the NN prediction model and results in an up to 7.3 percent increase in performance under a thermal constraint for mixed workloads compared to architecture-aware state-of-the-art (up to 20 percent increase for individual applications). This is achieved with a run-time overhead of less than 0.5 percent." @default.
• W3085005384 created "2020-09-21" @default.
• W3085005384 creator A5017470419 @default.
• W3085005384 creator A5049237676 @default.
• W3085005384 creator A5063508488 @default.
• W3085005384 creator A5067055700 @default.
• W3085005384 date "2020-01-01" @default.
• W3085005384 modified "2023-10-09" @default.
• W3085005384 title "Neural Network-based Performance Prediction for Task Migration on S-NUCA Many-Cores" @default.
• W3085005384 cites W1596277129 @default.
• W3085005384 cites W1965074384 @default.
• W3085005384 cites W1980072970 @default.
• W3085005384 cites W1985387589 @default.
• W3085005384 cites W2027829345 @default.
• W3085005384 cites W2033797066 @default.
• W3085005384 cites W2034062945 @default.
• W3085005384 cites W2049247044 @default.
• W3085005384 cites W2062360490 @default.
• W3085005384 cites W2079218824 @default.
• W3085005384 cites W2079219249 @default.
• W3085005384 cites W2102566970 @default.
• W3085005384 cites W2126372249 @default.
• W3085005384 cites W2129960401 @default.
• W3085005384 cites W2134633067 @default.
• W3085005384 cites W2161678417 @default.
• W3085005384 cites W2162082016 @default.
• W3085005384 cites W2169875292 @default.
• W3085005384 cites W2232834753 @default.
• W3085005384 cites W2342077212 @default.
• W3085005384 cites W2345737886 @default.
• W3085005384 cites W2415129417 @default.
• W3085005384 cites W2474840560 @default.
• W3085005384 cites W2562225561 @default.
• W3085005384 cites W2590858232 @default.
• W3085005384 cites W2749844496 @default.
• W3085005384 cites W2754640706 @default.
• W3085005384 cites W2770465907 @default.
• W3085005384 cites W2798972322 @default.
• W3085005384 cites W2898122376 @default.
• W3085005384 cites W2947723533 @default.
• W3085005384 cites W2963122961 @default.
• W3085005384 cites W2969250250 @default.
• W3085005384 cites W3042482718 @default.
• W3085005384 cites W4235431331 @default.
• W3085005384 cites W4235679005 @default.
• W3085005384 cites W4244536239 @default.
• W3085005384 cites W4248775335 @default.
• W3085005384 doi "https://doi.org/10.1109/tc.2020.3023022" @default.
• W3085005384 hasPublicationYear "2020" @default.
• W3085005384 type Work @default.
• W3085005384 sameAs 3085005384 @default.
• W3085005384 citedByCount "9" @default.
• W3085005384 countsByYear W30850053842021 @default.
• W3085005384 countsByYear W30850053842022 @default.
• W3085005384 countsByYear W30850053842023 @default.
• W3085005384 crossrefType "journal-article" @default.
• W3085005384 hasAuthorship W3085005384A5017470419 @default.
• W3085005384 hasAuthorship W3085005384A5049237676 @default.
• W3085005384 hasAuthorship W3085005384A5063508488 @default.
• W3085005384 hasAuthorship W3085005384A5067055700 @default.
• W3085005384 hasConcept C111919701 @default.
• W3085005384 hasConcept C115537543 @default.
• W3085005384 hasConcept C120314980 @default.
• W3085005384 hasConcept C121332964 @default.
• W3085005384 hasConcept C138101251 @default.
• W3085005384 hasConcept C149768029 @default.
• W3085005384 hasConcept C154945302 @default.
• W3085005384 hasConcept C162324750 @default.
• W3085005384 hasConcept C163258240 @default.
• W3085005384 hasConcept C173608175 @default.
• W3085005384 hasConcept C187736073 @default.
• W3085005384 hasConcept C2164484 @default.
• W3085005384 hasConcept C2780451532 @default.
• W3085005384 hasConcept C41008148 @default.
• W3085005384 hasConcept C50644808 @default.
• W3085005384 hasConcept C56685638 @default.
• W3085005384 hasConcept C62520636 @default.
• W3085005384 hasConcept C76155785 @default.
• W3085005384 hasConcept C78766204 @default.
• W3085005384 hasConcept C82876162 @default.
• W3085005384 hasConceptScore W3085005384C111919701 @default.
• W3085005384 hasConceptScore W3085005384C115537543 @default.
• W3085005384 hasConceptScore W3085005384C120314980 @default.
• W3085005384 hasConceptScore W3085005384C121332964 @default.
• W3085005384 hasConceptScore W3085005384C138101251 @default.
• W3085005384 hasConceptScore W3085005384C149768029 @default.
• W3085005384 hasConceptScore W3085005384C154945302 @default.
• W3085005384 hasConceptScore W3085005384C162324750 @default.
• W3085005384 hasConceptScore W3085005384C163258240 @default.
• W3085005384 hasConceptScore W3085005384C173608175 @default.
• W3085005384 hasConceptScore W3085005384C187736073 @default.
• W3085005384 hasConceptScore W3085005384C2164484 @default.
• W3085005384 hasConceptScore W3085005384C2780451532 @default.
• W3085005384 hasConceptScore W3085005384C41008148 @default.
• W3085005384 hasConceptScore W3085005384C50644808 @default.
• W3085005384 hasConceptScore W3085005384C56685638 @default.
• W3085005384 hasConceptScore W3085005384C62520636 @default.
• W3085005384 hasConceptScore W3085005384C76155785 @default.

• next