Placing pipeline stages on a Grid: Single path and multipath pipeline execution

In a Grid computing environment, several applications such as scientific data analysis and visualization are naturally computation and communication intensive. These applications can be decomposed into a sequence of pipeline stages which can be placed on different Grid nodes for concurrent execution...

Full description

Saved in:
Bibliographic Details
Main Authors: Ekasit Kijsipongse, Sudsanguan Ngamsuriyaroj
Other Authors: Mahidol University
Format: Article
Published: 2018
Subjects:
Online Access:https://repository.li.mahidol.ac.th/handle/123456789/29027
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Mahidol University
id th-mahidol.29027
record_format dspace
spelling th-mahidol.290272018-09-24T15:57:50Z Placing pipeline stages on a Grid: Single path and multipath pipeline execution Ekasit Kijsipongse Sudsanguan Ngamsuriyaroj Mahidol University Computer Science In a Grid computing environment, several applications such as scientific data analysis and visualization are naturally computation and communication intensive. These applications can be decomposed into a sequence of pipeline stages which can be placed on different Grid nodes for concurrent execution. Due to the aggregation of the computation and communication costs involved, finding the way to place such pipeline stages on a Grid in order to achieve the maximum application throughput becomes a challenging problem. This paper proposes a solution that considers both the pipeline placement and the data movement between stages. Specifically, we try to minimize the computation cost of the pipeline stages while preventing the communication overhead between successive stages from dominating the entire processing time. Our proposed solution consists of two novel methods. The first method is single path pipeline execution, which exploits only temporal parallelism, and the second method is multipath pipeline execution, which considers both temporal and spatial parallelism inherent in any pipeline applications. We evaluate our work in a simulated environment and also conduct a set of experiments in a real Grid computing system. When compared with the results from several traditional placement methods, our proposed methods give the highest throughput. © 2009 Elsevier B.V. All rights reserved. 2018-09-24T08:57:50Z 2018-09-24T08:57:50Z 2010-01-01 Article Future Generation Computer Systems. Vol.26, No.1 (2010), 50-62 10.1016/j.future.2009.06.005 0167739X 2-s2.0-69549138067 https://repository.li.mahidol.ac.th/handle/123456789/29027 Mahidol University SCOPUS https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=69549138067&origin=inward
institution Mahidol University
building Mahidol University Library
continent Asia
country Thailand
Thailand
content_provider Mahidol University Library
collection Mahidol University Institutional Repository
topic Computer Science
spellingShingle Computer Science
Ekasit Kijsipongse
Sudsanguan Ngamsuriyaroj
Placing pipeline stages on a Grid: Single path and multipath pipeline execution
description In a Grid computing environment, several applications such as scientific data analysis and visualization are naturally computation and communication intensive. These applications can be decomposed into a sequence of pipeline stages which can be placed on different Grid nodes for concurrent execution. Due to the aggregation of the computation and communication costs involved, finding the way to place such pipeline stages on a Grid in order to achieve the maximum application throughput becomes a challenging problem. This paper proposes a solution that considers both the pipeline placement and the data movement between stages. Specifically, we try to minimize the computation cost of the pipeline stages while preventing the communication overhead between successive stages from dominating the entire processing time. Our proposed solution consists of two novel methods. The first method is single path pipeline execution, which exploits only temporal parallelism, and the second method is multipath pipeline execution, which considers both temporal and spatial parallelism inherent in any pipeline applications. We evaluate our work in a simulated environment and also conduct a set of experiments in a real Grid computing system. When compared with the results from several traditional placement methods, our proposed methods give the highest throughput. © 2009 Elsevier B.V. All rights reserved.
author2 Mahidol University
author_facet Mahidol University
Ekasit Kijsipongse
Sudsanguan Ngamsuriyaroj
format Article
author Ekasit Kijsipongse
Sudsanguan Ngamsuriyaroj
author_sort Ekasit Kijsipongse
title Placing pipeline stages on a Grid: Single path and multipath pipeline execution
title_short Placing pipeline stages on a Grid: Single path and multipath pipeline execution
title_full Placing pipeline stages on a Grid: Single path and multipath pipeline execution
title_fullStr Placing pipeline stages on a Grid: Single path and multipath pipeline execution
title_full_unstemmed Placing pipeline stages on a Grid: Single path and multipath pipeline execution
title_sort placing pipeline stages on a grid: single path and multipath pipeline execution
publishDate 2018
url https://repository.li.mahidol.ac.th/handle/123456789/29027
_version_ 1763490106580140032