Dataflow partitioning and scheduling algorithms for WASMII, a virtual hardware

Atsushi Takayama, Yuichiro Shibata, Keisuke Iwai, Hideharu Amano

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Citations (Scopus)

Abstract

This paper presents a new dataflow graph partitioning algorithm for a data driven virtual hardware system called WASMII. The algorithm divides a dataflow graph into multiple subgraphs so as not to cause a deadlock. Then the subgraphs are translated into an FPGA configuration and executed on WASMII in a time-multiplexed manner. The experimental results show the proposed algorithms can achieve 13% to 39% improvement of execution performance compared to other existing graph partitioning algorithms at the most.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages685-694
Number of pages10
Volume1896
ISBN (Print)3540678999, 9783540678991
Publication statusPublished - 2000
Event10th International Conference on Field-Programmable Logic and Applications, FPL 2000 - Villach, Austria
Duration: 2000 Aug 272000 Aug 30

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1896
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other10th International Conference on Field-Programmable Logic and Applications, FPL 2000
CountryAustria
CityVillach
Period00/8/2700/8/30

Fingerprint

Scheduling algorithms
Data Flow
Scheduling Algorithm
Partitioning
Hardware
Graph Partitioning
Subgraph
Deadlock
Data-driven
Field Programmable Gate Array
Divides
Field programmable gate arrays (FPGA)
Configuration
Experimental Results
Graph in graph theory

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Takayama, A., Shibata, Y., Iwai, K., & Amano, H. (2000). Dataflow partitioning and scheduling algorithms for WASMII, a virtual hardware. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1896, pp. 685-694). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1896). Springer Verlag.

Dataflow partitioning and scheduling algorithms for WASMII, a virtual hardware. / Takayama, Atsushi; Shibata, Yuichiro; Iwai, Keisuke; Amano, Hideharu.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 1896 Springer Verlag, 2000. p. 685-694 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1896).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Takayama, A, Shibata, Y, Iwai, K & Amano, H 2000, Dataflow partitioning and scheduling algorithms for WASMII, a virtual hardware. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 1896, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1896, Springer Verlag, pp. 685-694, 10th International Conference on Field-Programmable Logic and Applications, FPL 2000, Villach, Austria, 00/8/27.
Takayama A, Shibata Y, Iwai K, Amano H. Dataflow partitioning and scheduling algorithms for WASMII, a virtual hardware. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 1896. Springer Verlag. 2000. p. 685-694. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Takayama, Atsushi ; Shibata, Yuichiro ; Iwai, Keisuke ; Amano, Hideharu. / Dataflow partitioning and scheduling algorithms for WASMII, a virtual hardware. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 1896 Springer Verlag, 2000. pp. 685-694 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{bd8ba68f737b4148ba6eef65c494ca96,
title = "Dataflow partitioning and scheduling algorithms for WASMII, a virtual hardware",
abstract = "This paper presents a new dataflow graph partitioning algorithm for a data driven virtual hardware system called WASMII. The algorithm divides a dataflow graph into multiple subgraphs so as not to cause a deadlock. Then the subgraphs are translated into an FPGA configuration and executed on WASMII in a time-multiplexed manner. The experimental results show the proposed algorithms can achieve 13{\%} to 39{\%} improvement of execution performance compared to other existing graph partitioning algorithms at the most.",
author = "Atsushi Takayama and Yuichiro Shibata and Keisuke Iwai and Hideharu Amano",
year = "2000",
language = "English",
isbn = "3540678999",
volume = "1896",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "685--694",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Dataflow partitioning and scheduling algorithms for WASMII, a virtual hardware

AU - Takayama, Atsushi

AU - Shibata, Yuichiro

AU - Iwai, Keisuke

AU - Amano, Hideharu

PY - 2000

Y1 - 2000

N2 - This paper presents a new dataflow graph partitioning algorithm for a data driven virtual hardware system called WASMII. The algorithm divides a dataflow graph into multiple subgraphs so as not to cause a deadlock. Then the subgraphs are translated into an FPGA configuration and executed on WASMII in a time-multiplexed manner. The experimental results show the proposed algorithms can achieve 13% to 39% improvement of execution performance compared to other existing graph partitioning algorithms at the most.

AB - This paper presents a new dataflow graph partitioning algorithm for a data driven virtual hardware system called WASMII. The algorithm divides a dataflow graph into multiple subgraphs so as not to cause a deadlock. Then the subgraphs are translated into an FPGA configuration and executed on WASMII in a time-multiplexed manner. The experimental results show the proposed algorithms can achieve 13% to 39% improvement of execution performance compared to other existing graph partitioning algorithms at the most.

UR - http://www.scopus.com/inward/record.url?scp=84947576966&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84947576966&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84947576966

SN - 3540678999

SN - 9783540678991

VL - 1896

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 685

EP - 694

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -