A dataflow-inspired CGRA for streaming applications

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)

Abstract

The herein presented research is motivated by the need for reconfigurable, flexible computing arrays targeted at streaming applications that contain a large degree of instruction-level parallelism. Such arrays are usually referred to as coarse-grained reconfigurable arrays (CGRAs). CGRAs are composed of small, reconfigurable cores that are interconnected to form a computing grid. When reviewing related work, it became clear to us that an intuitive, easy-to-use programming paradigm was missing. The vast majority of the available CGRAs use a C-based programming approach. In our opinion however, C is not the most obvious choice to program such systems, since C has originally been designed as a sequential language, and therefore lacks support for structural description of parallelism. As a consequence, the burden of extracting the structure of an implemented algorithm lies on the compiler. Here, we present a complete CGRA, consisting of an architecture and a programming language. Both the archi- tecture and the programming language are inspired by the principles found in dataflow. In particular, the firing rule, which means that a node fires as soon as all the required inputs are available. Furthermore, instead of developing a programming paradigm based on C, we decided to base our programming paradigm on the base of the functional programming language Haskell, that already has a notion of structure and the ability to describe instruction-level parallelism.
Original languageUndefined
Title of host publication23rd International Conference on Field Programmable Logic and Applications, FPL 2013
Place of PublicationLos Alamitos, CA, USA
PublisherIEEE Circuits & Systems Society
Pages1-2
Number of pages2
ISBN (Print)978-1-4799-0004-6
DOIs
Publication statusPublished - 2 Sep 2013
Event23rd International Conference on Field Programmable Logic and Applications, FPL 2013 - HF Ipanema Park, Porto, Portugal
Duration: 2 Sep 20134 Sep 2013
Conference number: 23

Publication series

Name
PublisherIEEE Circuits & Systems Society

Conference

Conference23rd International Conference on Field Programmable Logic and Applications, FPL 2013
Abbreviated titleFPL
CountryPortugal
CityPorto
Period2/09/134/09/13

Keywords

  • EWI-23892
  • IR-87612
  • METIS-300119

Cite this

Niedermeier, A., Kuper, J., & Smit, G. J. M. (2013). A dataflow-inspired CGRA for streaming applications. In 23rd International Conference on Field Programmable Logic and Applications, FPL 2013 (pp. 1-2). Los Alamitos, CA, USA: IEEE Circuits & Systems Society. https://doi.org/10.1109/FPL.2013.6645609
Niedermeier, A. ; Kuper, Jan ; Smit, Gerardus Johannes Maria. / A dataflow-inspired CGRA for streaming applications. 23rd International Conference on Field Programmable Logic and Applications, FPL 2013. Los Alamitos, CA, USA : IEEE Circuits & Systems Society, 2013. pp. 1-2
@inproceedings{af30b110bf3d4313a9cdf6eef83ee8d8,
title = "A dataflow-inspired CGRA for streaming applications",
abstract = "The herein presented research is motivated by the need for reconfigurable, flexible computing arrays targeted at streaming applications that contain a large degree of instruction-level parallelism. Such arrays are usually referred to as coarse-grained reconfigurable arrays (CGRAs). CGRAs are composed of small, reconfigurable cores that are interconnected to form a computing grid. When reviewing related work, it became clear to us that an intuitive, easy-to-use programming paradigm was missing. The vast majority of the available CGRAs use a C-based programming approach. In our opinion however, C is not the most obvious choice to program such systems, since C has originally been designed as a sequential language, and therefore lacks support for structural description of parallelism. As a consequence, the burden of extracting the structure of an implemented algorithm lies on the compiler. Here, we present a complete CGRA, consisting of an architecture and a programming language. Both the archi- tecture and the programming language are inspired by the principles found in dataflow. In particular, the firing rule, which means that a node fires as soon as all the required inputs are available. Furthermore, instead of developing a programming paradigm based on C, we decided to base our programming paradigm on the base of the functional programming language Haskell, that already has a notion of structure and the ability to describe instruction-level parallelism.",
keywords = "EWI-23892, IR-87612, METIS-300119",
author = "A. Niedermeier and Jan Kuper and Smit, {Gerardus Johannes Maria}",
note = "eemcs-eprint-23892",
year = "2013",
month = "9",
day = "2",
doi = "10.1109/FPL.2013.6645609",
language = "Undefined",
isbn = "978-1-4799-0004-6",
publisher = "IEEE Circuits & Systems Society",
pages = "1--2",
booktitle = "23rd International Conference on Field Programmable Logic and Applications, FPL 2013",

}

Niedermeier, A, Kuper, J & Smit, GJM 2013, A dataflow-inspired CGRA for streaming applications. in 23rd International Conference on Field Programmable Logic and Applications, FPL 2013. IEEE Circuits & Systems Society, Los Alamitos, CA, USA, pp. 1-2, 23rd International Conference on Field Programmable Logic and Applications, FPL 2013, Porto, Portugal, 2/09/13. https://doi.org/10.1109/FPL.2013.6645609

A dataflow-inspired CGRA for streaming applications. / Niedermeier, A.; Kuper, Jan; Smit, Gerardus Johannes Maria.

23rd International Conference on Field Programmable Logic and Applications, FPL 2013. Los Alamitos, CA, USA : IEEE Circuits & Systems Society, 2013. p. 1-2.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - A dataflow-inspired CGRA for streaming applications

AU - Niedermeier, A.

AU - Kuper, Jan

AU - Smit, Gerardus Johannes Maria

N1 - eemcs-eprint-23892

PY - 2013/9/2

Y1 - 2013/9/2

N2 - The herein presented research is motivated by the need for reconfigurable, flexible computing arrays targeted at streaming applications that contain a large degree of instruction-level parallelism. Such arrays are usually referred to as coarse-grained reconfigurable arrays (CGRAs). CGRAs are composed of small, reconfigurable cores that are interconnected to form a computing grid. When reviewing related work, it became clear to us that an intuitive, easy-to-use programming paradigm was missing. The vast majority of the available CGRAs use a C-based programming approach. In our opinion however, C is not the most obvious choice to program such systems, since C has originally been designed as a sequential language, and therefore lacks support for structural description of parallelism. As a consequence, the burden of extracting the structure of an implemented algorithm lies on the compiler. Here, we present a complete CGRA, consisting of an architecture and a programming language. Both the archi- tecture and the programming language are inspired by the principles found in dataflow. In particular, the firing rule, which means that a node fires as soon as all the required inputs are available. Furthermore, instead of developing a programming paradigm based on C, we decided to base our programming paradigm on the base of the functional programming language Haskell, that already has a notion of structure and the ability to describe instruction-level parallelism.

AB - The herein presented research is motivated by the need for reconfigurable, flexible computing arrays targeted at streaming applications that contain a large degree of instruction-level parallelism. Such arrays are usually referred to as coarse-grained reconfigurable arrays (CGRAs). CGRAs are composed of small, reconfigurable cores that are interconnected to form a computing grid. When reviewing related work, it became clear to us that an intuitive, easy-to-use programming paradigm was missing. The vast majority of the available CGRAs use a C-based programming approach. In our opinion however, C is not the most obvious choice to program such systems, since C has originally been designed as a sequential language, and therefore lacks support for structural description of parallelism. As a consequence, the burden of extracting the structure of an implemented algorithm lies on the compiler. Here, we present a complete CGRA, consisting of an architecture and a programming language. Both the archi- tecture and the programming language are inspired by the principles found in dataflow. In particular, the firing rule, which means that a node fires as soon as all the required inputs are available. Furthermore, instead of developing a programming paradigm based on C, we decided to base our programming paradigm on the base of the functional programming language Haskell, that already has a notion of structure and the ability to describe instruction-level parallelism.

KW - EWI-23892

KW - IR-87612

KW - METIS-300119

U2 - 10.1109/FPL.2013.6645609

DO - 10.1109/FPL.2013.6645609

M3 - Conference contribution

SN - 978-1-4799-0004-6

SP - 1

EP - 2

BT - 23rd International Conference on Field Programmable Logic and Applications, FPL 2013

PB - IEEE Circuits & Systems Society

CY - Los Alamitos, CA, USA

ER -

Niedermeier A, Kuper J, Smit GJM. A dataflow-inspired CGRA for streaming applications. In 23rd International Conference on Field Programmable Logic and Applications, FPL 2013. Los Alamitos, CA, USA: IEEE Circuits & Systems Society. 2013. p. 1-2 https://doi.org/10.1109/FPL.2013.6645609