Matches in SemOpenAlex for { <https://semopenalex.org/work/W2914005648> ?p ?o ?g. }
- W2914005648 endingPage "189" @default.
- W2914005648 startingPage "178" @default.
- W2914005648 abstract "Pipelining is a well-known technique to overlap loop iterations by partitioning the loop body into a sequence of stages. A large class of programs can be expressed as linear pipelines if data dependences only flow from earlier to later stages. Various pipelining techniques have been explored but reconciling load-balancing and efficient execution is still a challenge for two main reasons. First, partitioning of the loop body into stages that lead to load-balancing may depend on the data set as well as system properties, e.g., number of cores. Second, the configuration of the runtime system is far from obvious. In this article, we present Pipelight, a technique that achieves load-balancing for linear pipelines. Pipelight relies on a way of mapping stages onto threads that simplifies partitioning and enables the design of a lightweight algorithm for dynamic scheduling. Furthermore, Pipelight introduces a concurrent data structure that exploits the properties of data dependences presented in linear pipelines to provide efficient communication and synchronization. This data structure simplifies the configuration of the runtime system and makes Pipelight a practical solution. The evaluation on a 44-core system shows the efficiency of Pipelight for a set of programs selected from widely-used collections. Although Pipelight simplifies parallelization of linear pipelines, it performs similarly to the most efficient properly configured state-of-the-art technique. The price paid for the benefits of Pipelight is additional overhead for fine-grained loops. However, this overhead can be amortized successfully with chunking. To make Pipelight a promising solution, we propose a directive-based transformation for Pipelight, which is developed in a prototype source-to-source compiler. Consequently, Pipelight is an efficient and practical solution to achieve load-balancing for fine-grained linear pipelines." @default.
- W2914005648 created "2019-02-21" @default.
- W2914005648 creator A5026061564 @default.
- W2914005648 creator A5028216967 @default.
- W2914005648 date "2019-07-01" @default.
- W2914005648 modified "2023-10-18" @default.
- W2914005648 title "Load-balancing for load-imbalanced fine-grained linear pipelines" @default.
- W2914005648 cites W1555823978 @default.
- W2914005648 cites W1964960933 @default.
- W2914005648 cites W1969240521 @default.
- W2914005648 cites W1976397984 @default.
- W2914005648 cites W1982812416 @default.
- W2914005648 cites W1990199656 @default.
- W2914005648 cites W2006059438 @default.
- W2914005648 cites W2009964586 @default.
- W2914005648 cites W2034209740 @default.
- W2914005648 cites W2040523061 @default.
- W2914005648 cites W2044872254 @default.
- W2914005648 cites W2046188666 @default.
- W2914005648 cites W2061153463 @default.
- W2914005648 cites W2072725684 @default.
- W2914005648 cites W2082056472 @default.
- W2914005648 cites W2086779607 @default.
- W2914005648 cites W2090408221 @default.
- W2914005648 cites W2094020960 @default.
- W2914005648 cites W2122698306 @default.
- W2914005648 cites W2128294826 @default.
- W2914005648 cites W2143592587 @default.
- W2914005648 cites W2143746374 @default.
- W2914005648 cites W2145406052 @default.
- W2914005648 cites W2147536281 @default.
- W2914005648 cites W2155312437 @default.
- W2914005648 cites W2167089275 @default.
- W2914005648 cites W2169875292 @default.
- W2914005648 cites W2171546589 @default.
- W2914005648 cites W2176686728 @default.
- W2914005648 cites W2253690730 @default.
- W2914005648 cites W2789351849 @default.
- W2914005648 cites W2793755249 @default.
- W2914005648 cites W79448102 @default.
- W2914005648 doi "https://doi.org/10.1016/j.parco.2019.01.002" @default.
- W2914005648 hasPublicationYear "2019" @default.
- W2914005648 type Work @default.
- W2914005648 sameAs 2914005648 @default.
- W2914005648 citedByCount "0" @default.
- W2914005648 crossrefType "journal-article" @default.
- W2914005648 hasAuthorship W2914005648A5026061564 @default.
- W2914005648 hasAuthorship W2914005648A5028216967 @default.
- W2914005648 hasBestOaLocation W29140056481 @default.
- W2914005648 hasConcept C111919701 @default.
- W2914005648 hasConcept C120314980 @default.
- W2914005648 hasConcept C127413603 @default.
- W2914005648 hasConcept C138959212 @default.
- W2914005648 hasConcept C162324750 @default.
- W2914005648 hasConcept C165696696 @default.
- W2914005648 hasConcept C173608175 @default.
- W2914005648 hasConcept C175309249 @default.
- W2914005648 hasConcept C187691185 @default.
- W2914005648 hasConcept C199360897 @default.
- W2914005648 hasConcept C206729178 @default.
- W2914005648 hasConcept C21547014 @default.
- W2914005648 hasConcept C2524010 @default.
- W2914005648 hasConcept C2779960059 @default.
- W2914005648 hasConcept C33923547 @default.
- W2914005648 hasConcept C38652104 @default.
- W2914005648 hasConcept C41008148 @default.
- W2914005648 hasConcept C43521106 @default.
- W2914005648 hasConcept C489000 @default.
- W2914005648 hasConcept C77088390 @default.
- W2914005648 hasConcept C87717796 @default.
- W2914005648 hasConceptScore W2914005648C111919701 @default.
- W2914005648 hasConceptScore W2914005648C120314980 @default.
- W2914005648 hasConceptScore W2914005648C127413603 @default.
- W2914005648 hasConceptScore W2914005648C138959212 @default.
- W2914005648 hasConceptScore W2914005648C162324750 @default.
- W2914005648 hasConceptScore W2914005648C165696696 @default.
- W2914005648 hasConceptScore W2914005648C173608175 @default.
- W2914005648 hasConceptScore W2914005648C175309249 @default.
- W2914005648 hasConceptScore W2914005648C187691185 @default.
- W2914005648 hasConceptScore W2914005648C199360897 @default.
- W2914005648 hasConceptScore W2914005648C206729178 @default.
- W2914005648 hasConceptScore W2914005648C21547014 @default.
- W2914005648 hasConceptScore W2914005648C2524010 @default.
- W2914005648 hasConceptScore W2914005648C2779960059 @default.
- W2914005648 hasConceptScore W2914005648C33923547 @default.
- W2914005648 hasConceptScore W2914005648C38652104 @default.
- W2914005648 hasConceptScore W2914005648C41008148 @default.
- W2914005648 hasConceptScore W2914005648C43521106 @default.
- W2914005648 hasConceptScore W2914005648C489000 @default.
- W2914005648 hasConceptScore W2914005648C77088390 @default.
- W2914005648 hasConceptScore W2914005648C87717796 @default.
- W2914005648 hasFunder F4320320924 @default.
- W2914005648 hasLocation W29140056481 @default.
- W2914005648 hasLocation W29140056482 @default.
- W2914005648 hasOpenAccess W2914005648 @default.
- W2914005648 hasPrimaryLocation W29140056481 @default.
- W2914005648 hasRelatedWork W1513409726 @default.
- W2914005648 hasRelatedWork W1572523360 @default.