Matches in SemOpenAlex for { <https://semopenalex.org/work/W2039727401> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W2039727401 abstract "Graphics processing units (GPUs) are increasingly utilized as throughput engines in the modern computer systems. GPUs rely on fast context switching between thousands of threads to hide long latency operations, however, they still stall due to the memory operations. To minimize the stalls, memory operations should be overlapped with other operations as much as possible to maximize memory-level parallelism (MLP). In this paper, we propose Earliest Load First (ELF) warp scheduling, which maximizes the MLP by giving higher priority to the warps that have the fewest instructions to the next memory load. ELF utilizes the same warp priority for the fetch scheduling so that both are coordinated. We also show that ELF reveals its full benefits when there are fewer memory conflicts and fetch stalls. Evaluations show that ELF can improve the performance by 4.1% and achieve total improvement of 11.9% when used with other techniques over commonly-used greedy-then-oldest scheduling." @default.
- W2039727401 created "2016-06-24" @default.
- W2039727401 creator A5002075773 @default.
- W2039727401 creator A5023758504 @default.
- W2039727401 creator A5086980677 @default.
- W2039727401 date "2015-11-15" @default.
- W2039727401 modified "2023-10-16" @default.
- W2039727401 title "ELF" @default.
- W2039727401 cites W1979527452 @default.
- W2039727401 cites W2001784723 @default.
- W2039727401 cites W2013281054 @default.
- W2039727401 cites W2047060659 @default.
- W2039727401 cites W2053744175 @default.
- W2039727401 cites W2067441262 @default.
- W2039727401 cites W2079038734 @default.
- W2039727401 cites W2080592089 @default.
- W2039727401 cites W2090584832 @default.
- W2039727401 cites W2095804977 @default.
- W2039727401 cites W2096661534 @default.
- W2039727401 cites W2098505406 @default.
- W2039727401 cites W2122422937 @default.
- W2039727401 cites W2142444503 @default.
- W2039727401 cites W2143749395 @default.
- W2039727401 cites W2156355519 @default.
- W2039727401 cites W2166918318 @default.
- W2039727401 cites W2207094710 @default.
- W2039727401 cites W3144376511 @default.
- W2039727401 cites W4232096869 @default.
- W2039727401 doi "https://doi.org/10.1145/2807591.2807598" @default.
- W2039727401 hasPublicationYear "2015" @default.
- W2039727401 type Work @default.
- W2039727401 sameAs 2039727401 @default.
- W2039727401 citedByCount "12" @default.
- W2039727401 countsByYear W20397274012016 @default.
- W2039727401 countsByYear W20397274012017 @default.
- W2039727401 countsByYear W20397274012018 @default.
- W2039727401 countsByYear W20397274012019 @default.
- W2039727401 countsByYear W20397274012021 @default.
- W2039727401 crossrefType "proceedings-article" @default.
- W2039727401 hasAuthorship W2039727401A5002075773 @default.
- W2039727401 hasAuthorship W2039727401A5023758504 @default.
- W2039727401 hasAuthorship W2039727401A5086980677 @default.
- W2039727401 hasBestOaLocation W20397274011 @default.
- W2039727401 hasConcept C111368507 @default.
- W2039727401 hasConcept C120314980 @default.
- W2039727401 hasConcept C127313418 @default.
- W2039727401 hasConcept C127413603 @default.
- W2039727401 hasConcept C131918245 @default.
- W2039727401 hasConcept C149635348 @default.
- W2039727401 hasConcept C173608175 @default.
- W2039727401 hasConcept C206729178 @default.
- W2039727401 hasConcept C21547014 @default.
- W2039727401 hasConcept C41008148 @default.
- W2039727401 hasConcept C53833338 @default.
- W2039727401 hasConcept C76155785 @default.
- W2039727401 hasConcept C82876162 @default.
- W2039727401 hasConceptScore W2039727401C111368507 @default.
- W2039727401 hasConceptScore W2039727401C120314980 @default.
- W2039727401 hasConceptScore W2039727401C127313418 @default.
- W2039727401 hasConceptScore W2039727401C127413603 @default.
- W2039727401 hasConceptScore W2039727401C131918245 @default.
- W2039727401 hasConceptScore W2039727401C149635348 @default.
- W2039727401 hasConceptScore W2039727401C173608175 @default.
- W2039727401 hasConceptScore W2039727401C206729178 @default.
- W2039727401 hasConceptScore W2039727401C21547014 @default.
- W2039727401 hasConceptScore W2039727401C41008148 @default.
- W2039727401 hasConceptScore W2039727401C53833338 @default.
- W2039727401 hasConceptScore W2039727401C76155785 @default.
- W2039727401 hasConceptScore W2039727401C82876162 @default.
- W2039727401 hasFunder F4320306076 @default.
- W2039727401 hasFunder F4320332180 @default.
- W2039727401 hasLocation W20397274011 @default.
- W2039727401 hasOpenAccess W2039727401 @default.
- W2039727401 hasPrimaryLocation W20397274011 @default.
- W2039727401 hasRelatedWork W1523834180 @default.
- W2039727401 hasRelatedWork W1882733036 @default.
- W2039727401 hasRelatedWork W1992741870 @default.
- W2039727401 hasRelatedWork W2028089209 @default.
- W2039727401 hasRelatedWork W2104603305 @default.
- W2039727401 hasRelatedWork W2139514495 @default.
- W2039727401 hasRelatedWork W2160425906 @default.
- W2039727401 hasRelatedWork W2298102683 @default.
- W2039727401 hasRelatedWork W2546696010 @default.
- W2039727401 hasRelatedWork W3140076381 @default.
- W2039727401 isParatext "false" @default.
- W2039727401 isRetracted "false" @default.
- W2039727401 magId "2039727401" @default.
- W2039727401 workType "article" @default.