2009 15th International Conference on Parallel and Distributed Systems 2009
DOI: 10.1109/icpads.2009.143
|View full text |Cite
|
Sign up to set email alerts
|

Exploiting TLS Parallelism at Multiple Loop-Nest Levels

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
1
0
1

Year Published

2013
2013
2013
2013

Publication Types

Select...
1
1

Relationship

0
2

Authors

Journals

citations
Cited by 2 publications
(2 citation statements)
references
References 11 publications
0
1
0
1
Order By: Relevance
“…Meanwhile, nested data processing has been explored by the database community for several decades, such as Volcano [Graefe 1994] data-flow query processing systems. In the compiler community, Packirisamy and Zhai [2009] design a new algorithm to exploit parallelism for programs with nested loops by statically mapping cores to different levels of loop-nests. We observed that the Map-Reduce programming model also could be paralleled in two dimensions and it could be efficiently implemented on multicore architecture.…”
Section: Nested Data Parallelismmentioning
confidence: 99%
“…Meanwhile, nested data processing has been explored by the database community for several decades, such as Volcano [Graefe 1994] data-flow query processing systems. In the compiler community, Packirisamy and Zhai [2009] design a new algorithm to exploit parallelism for programs with nested loops by statically mapping cores to different levels of loop-nests. We observed that the Map-Reduce programming model also could be paralleled in two dimensions and it could be efficiently implemented on multicore architecture.…”
Section: Nested Data Parallelismmentioning
confidence: 99%
“…TLS, já citada anteriormente, é um bom exemplo deste tipo de heurística. Em especial, é possível ver em [39] o uso de TLS para permitir a laços aninhados o usufruto do potencial de processadores multicore. A implementação apresentada neste artigo não exigia modificações no hardware atual, mas obteve cerca de 15% de ganho, na média, no tempo de execução dos programas testados.…”
Section: Capítulo 2 Revisão Bibliográficaunclassified