Proceedings of the 7th Workshop on Declarative Aspects and Applications of Multicore Programming 2012
DOI: 10.1145/2103736.2103740
|View full text |Cite
|
Sign up to set email alerts
|

Expressive array constructs in an embedded GPU kernel programming language

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
54
0

Year Published

2013
2013
2019
2019

Publication Types

Select...
5
2
2

Relationship

0
9

Authors

Journals

citations
Cited by 60 publications
(54 citation statements)
references
References 11 publications
0
54
0
Order By: Relevance
“…These include our own Skel library [7], which is currently the only available skeleton library for Erlang. A recent alternative approach is to define an embedded domain-specific language (EDSL) [12,22], This can provide a structured approach to parallelism, similar to our approach using skeletons. Skeletons, however, have the advantage of representing language-independent patterns of parallelism, that can more easily be transferred to other language settings.…”
Section: Related Workmentioning
confidence: 99%
“…These include our own Skel library [7], which is currently the only available skeleton library for Erlang. A recent alternative approach is to define an embedded domain-specific language (EDSL) [12,22], This can provide a structured approach to parallelism, similar to our approach using skeletons. Skeletons, however, have the advantage of representing language-independent patterns of parallelism, that can more easily be transferred to other language settings.…”
Section: Related Workmentioning
confidence: 99%
“…There are also tagless-final DSLs [6], where there is no early commitment to a specific embedding, but instead class overloading in used to specify the API. Several Haskell DSLs, including Feldspar [37] and Obsidian [8], reify the structure of monadic code [40,41] to generate external imperative code. Other DSLs, including CoPilot [38] and Ivory [24], use a shallow embedding of statements to capture the monadic structure.…”
Section: Other Related Workmentioning
confidence: 99%
“…One of these types, Pull arrays, were already presented in section 3.2. The other type is Push arrays, a notion originally introduced in [5]. Push arrays shares many significant properties with Pull arrays: they can be fused just as easily, are efficiently parallelizeable, and have a Functor instance.…”
Section: Push Arraysmentioning
confidence: 99%
“…We have included Push arrays [5] in our implementation. The result is a simpler implementation of many array operations including stencil computations and although the user of our library must now use two different types of arrays we consider the resulting API to be easier to use.…”
Section: Introductionmentioning
confidence: 99%