A dataflow language with object-based extension and its implementation on a commercially available parallel machine

Shigeru Kusakabe, Taku Nagai, Yoshihiro Yamashita, Rin-Ichiro Taniguchi, Makoto Amamiya

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

5 Citations (Scopus)

Abstract

We propose a massively parallel programming language, called "V," which would minimize the difficulties in writing massively parallel programs. To abstract away the timing problem in writing parallel programs, we based our work on a dataflow-based functional programming language. Then, we extended the base language with an object-based abstraction, called "agent," to write parallel entities which contain their own states and can communicate with each other. In addition to connecting agents explicitly, an abstraction of ensembles of agents on a predefined topology description, called "field," is introduced in order to write a massively parallel program which naturally reflects the structure of a problem. In this paper, we also discuss the implementation of V, including a preliminary evaluation, on a commercially available distributed-memory parallel machine, Fujitsu AP1000. Although it is easy to extract parallelism at various levels, including finegrained parallelism, from V programs, it is difficult to run V programs efficiently on stock parallel machines. Nevertheless, we tried to implement V on a stock parallel machine, since our aim is to develop a language that would be practical, not only on special fine-grained machines, but also stock machines.

Original languageEnglish
Title of host publicationProceedings of the 9th International Conference on Supercomputing, ICS 1995
PublisherAssociation for Computing Machinery
Pages308-317
Number of pages10
ISBN (Electronic)0897917286
DOIs
Publication statusPublished - Jul 3 1995
Event9th International Conference on Supercomputing, ICS 1995 - Barcelona, Spain
Duration: Jul 3 1995Jul 7 1995

Publication series

NameProceedings of the International Conference on Supercomputing
VolumePart F129361

Other

Other9th International Conference on Supercomputing, ICS 1995
CountrySpain
CityBarcelona
Period7/3/957/7/95

Fingerprint

Computer programming languages
Functional programming
Parallel programming
Topology
Data storage equipment

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Cite this

Kusakabe, S., Nagai, T., Yamashita, Y., Taniguchi, R-I., & Amamiya, M. (1995). A dataflow language with object-based extension and its implementation on a commercially available parallel machine. In Proceedings of the 9th International Conference on Supercomputing, ICS 1995 (pp. 308-317). (Proceedings of the International Conference on Supercomputing; Vol. Part F129361). Association for Computing Machinery. https://doi.org/10.1145/224538.224575

A dataflow language with object-based extension and its implementation on a commercially available parallel machine. / Kusakabe, Shigeru; Nagai, Taku; Yamashita, Yoshihiro; Taniguchi, Rin-Ichiro; Amamiya, Makoto.

Proceedings of the 9th International Conference on Supercomputing, ICS 1995. Association for Computing Machinery, 1995. p. 308-317 (Proceedings of the International Conference on Supercomputing; Vol. Part F129361).

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

Kusakabe, S, Nagai, T, Yamashita, Y, Taniguchi, R-I & Amamiya, M 1995, A dataflow language with object-based extension and its implementation on a commercially available parallel machine. in Proceedings of the 9th International Conference on Supercomputing, ICS 1995. Proceedings of the International Conference on Supercomputing, vol. Part F129361, Association for Computing Machinery, pp. 308-317, 9th International Conference on Supercomputing, ICS 1995, Barcelona, Spain, 7/3/95. https://doi.org/10.1145/224538.224575
Kusakabe S, Nagai T, Yamashita Y, Taniguchi R-I, Amamiya M. A dataflow language with object-based extension and its implementation on a commercially available parallel machine. In Proceedings of the 9th International Conference on Supercomputing, ICS 1995. Association for Computing Machinery. 1995. p. 308-317. (Proceedings of the International Conference on Supercomputing). https://doi.org/10.1145/224538.224575
Kusakabe, Shigeru ; Nagai, Taku ; Yamashita, Yoshihiro ; Taniguchi, Rin-Ichiro ; Amamiya, Makoto. / A dataflow language with object-based extension and its implementation on a commercially available parallel machine. Proceedings of the 9th International Conference on Supercomputing, ICS 1995. Association for Computing Machinery, 1995. pp. 308-317 (Proceedings of the International Conference on Supercomputing).
@inproceedings{c7f15d684c64410a97040754a745ee3c,
title = "A dataflow language with object-based extension and its implementation on a commercially available parallel machine",
abstract = "We propose a massively parallel programming language, called {"}V,{"} which would minimize the difficulties in writing massively parallel programs. To abstract away the timing problem in writing parallel programs, we based our work on a dataflow-based functional programming language. Then, we extended the base language with an object-based abstraction, called {"}agent,{"} to write parallel entities which contain their own states and can communicate with each other. In addition to connecting agents explicitly, an abstraction of ensembles of agents on a predefined topology description, called {"}field,{"} is introduced in order to write a massively parallel program which naturally reflects the structure of a problem. In this paper, we also discuss the implementation of V, including a preliminary evaluation, on a commercially available distributed-memory parallel machine, Fujitsu AP1000. Although it is easy to extract parallelism at various levels, including finegrained parallelism, from V programs, it is difficult to run V programs efficiently on stock parallel machines. Nevertheless, we tried to implement V on a stock parallel machine, since our aim is to develop a language that would be practical, not only on special fine-grained machines, but also stock machines.",
author = "Shigeru Kusakabe and Taku Nagai and Yoshihiro Yamashita and Rin-Ichiro Taniguchi and Makoto Amamiya",
year = "1995",
month = "7",
day = "3",
doi = "10.1145/224538.224575",
language = "English",
series = "Proceedings of the International Conference on Supercomputing",
publisher = "Association for Computing Machinery",
pages = "308--317",
booktitle = "Proceedings of the 9th International Conference on Supercomputing, ICS 1995",

}

TY - GEN

T1 - A dataflow language with object-based extension and its implementation on a commercially available parallel machine

AU - Kusakabe, Shigeru

AU - Nagai, Taku

AU - Yamashita, Yoshihiro

AU - Taniguchi, Rin-Ichiro

AU - Amamiya, Makoto

PY - 1995/7/3

Y1 - 1995/7/3

N2 - We propose a massively parallel programming language, called "V," which would minimize the difficulties in writing massively parallel programs. To abstract away the timing problem in writing parallel programs, we based our work on a dataflow-based functional programming language. Then, we extended the base language with an object-based abstraction, called "agent," to write parallel entities which contain their own states and can communicate with each other. In addition to connecting agents explicitly, an abstraction of ensembles of agents on a predefined topology description, called "field," is introduced in order to write a massively parallel program which naturally reflects the structure of a problem. In this paper, we also discuss the implementation of V, including a preliminary evaluation, on a commercially available distributed-memory parallel machine, Fujitsu AP1000. Although it is easy to extract parallelism at various levels, including finegrained parallelism, from V programs, it is difficult to run V programs efficiently on stock parallel machines. Nevertheless, we tried to implement V on a stock parallel machine, since our aim is to develop a language that would be practical, not only on special fine-grained machines, but also stock machines.

AB - We propose a massively parallel programming language, called "V," which would minimize the difficulties in writing massively parallel programs. To abstract away the timing problem in writing parallel programs, we based our work on a dataflow-based functional programming language. Then, we extended the base language with an object-based abstraction, called "agent," to write parallel entities which contain their own states and can communicate with each other. In addition to connecting agents explicitly, an abstraction of ensembles of agents on a predefined topology description, called "field," is introduced in order to write a massively parallel program which naturally reflects the structure of a problem. In this paper, we also discuss the implementation of V, including a preliminary evaluation, on a commercially available distributed-memory parallel machine, Fujitsu AP1000. Although it is easy to extract parallelism at various levels, including finegrained parallelism, from V programs, it is difficult to run V programs efficiently on stock parallel machines. Nevertheless, we tried to implement V on a stock parallel machine, since our aim is to develop a language that would be practical, not only on special fine-grained machines, but also stock machines.

UR - http://www.scopus.com/inward/record.url?scp=0029206398&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029206398&partnerID=8YFLogxK

U2 - 10.1145/224538.224575

DO - 10.1145/224538.224575

M3 - Conference contribution

AN - SCOPUS:0029206398

T3 - Proceedings of the International Conference on Supercomputing

SP - 308

EP - 317

BT - Proceedings of the 9th International Conference on Supercomputing, ICS 1995

PB - Association for Computing Machinery

ER -