Workflow-based comparison of two distributed computing infrastructures

Johan Montagnat; Tristan Glatard; Damien Reimert; Ketan Maheshwari; Eddy Caron; Frédéric Desprez

doi:10.1109/WORKS.2010.5671856

Workflow-based comparison of two distributed computing infrastructures

Johan Montagnat, Tristan Glatard, Damien Reimert, Ketan Maheshwari, Eddy Caron, Frédéric Desprez

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

10 Scopus citations

Abstract

Porting applications to Distributed Computing Infrastructures (DCIs) is eased by the use of workflow abstractions. Yet, estimating the impact of the execution DCI on application performance is difficult due to the heterogeneity of the resources available, middleware and operation models. This paper describes a workflow-based experimental method to acquire objective performance comparison criterions when dealing with completely different DCIs. Experiments were conducted on the European EGI and the French Grid'5000 infrastructures to highlight raw performance variations and identify their causes. The results obtained also show that it is possible to conduct experiments on a production infrastructure with similar reproducibility as on an experimental platform.

Original language	English
Title of host publication	2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010
DOIs	https://doi.org/10.1109/WORKS.2010.5671856
State	Published - 2010
Externally published	Yes
Event	2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010 - New Orleans, LA, United States Duration: Nov 14 2010 → Nov 14 2010

Publication series

Name	2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010

Conference

Conference	2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010
Country/Territory	United States
City	New Orleans, LA
Period	11/14/10 → 11/14/10

Funding

Funders	Funder number
Seventh Framework Programme	261323, 261585

Keywords

C.4 [computer systems organization]
Distributed computing infrastructures
Performance
Performance of systems
Workflows

Access to Document

10.1109/WORKS.2010.5671856

Cite this

Montagnat, J., Glatard, T., Reimert, D., Maheshwari, K., Caron, E., & Desprez, F. (2010). Workflow-based comparison of two distributed computing infrastructures. In 2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010 Article 5671856 (2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010). https://doi.org/10.1109/WORKS.2010.5671856

@inproceedings{79012af6569549b8bbc7a723ce3b0b05,

title = "Workflow-based comparison of two distributed computing infrastructures",

abstract = "Porting applications to Distributed Computing Infrastructures (DCIs) is eased by the use of workflow abstractions. Yet, estimating the impact of the execution DCI on application performance is difficult due to the heterogeneity of the resources available, middleware and operation models. This paper describes a workflow-based experimental method to acquire objective performance comparison criterions when dealing with completely different DCIs. Experiments were conducted on the European EGI and the French Grid'5000 infrastructures to highlight raw performance variations and identify their causes. The results obtained also show that it is possible to conduct experiments on a production infrastructure with similar reproducibility as on an experimental platform.",

keywords = "C.4 [computer systems organization], Distributed computing infrastructures, Performance, Performance of systems, Workflows",

author = "Johan Montagnat and Tristan Glatard and Damien Reimert and Ketan Maheshwari and Eddy Caron and Fr{\'e}d{\'e}ric Desprez",

year = "2010",

doi = "10.1109/WORKS.2010.5671856",

language = "English",

isbn = "9781424489893",

series = "2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010",

booktitle = "2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010",

note = "2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010 ; Conference date: 14-11-2010 Through 14-11-2010",

}

Montagnat, J, Glatard, T, Reimert, D, Maheshwari, K, Caron, E & Desprez, F 2010, Workflow-based comparison of two distributed computing infrastructures. in 2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010., 5671856, 2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010, 2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010, New Orleans, LA, United States, 11/14/10. https://doi.org/10.1109/WORKS.2010.5671856

Workflow-based comparison of two distributed computing infrastructures. / Montagnat, Johan; Glatard, Tristan; Reimert, Damien et al.
2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010. 2010. 5671856 (2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Workflow-based comparison of two distributed computing infrastructures

AU - Montagnat, Johan

AU - Glatard, Tristan

AU - Reimert, Damien

AU - Maheshwari, Ketan

AU - Caron, Eddy

AU - Desprez, Frédéric

PY - 2010

Y1 - 2010

N2 - Porting applications to Distributed Computing Infrastructures (DCIs) is eased by the use of workflow abstractions. Yet, estimating the impact of the execution DCI on application performance is difficult due to the heterogeneity of the resources available, middleware and operation models. This paper describes a workflow-based experimental method to acquire objective performance comparison criterions when dealing with completely different DCIs. Experiments were conducted on the European EGI and the French Grid'5000 infrastructures to highlight raw performance variations and identify their causes. The results obtained also show that it is possible to conduct experiments on a production infrastructure with similar reproducibility as on an experimental platform.

AB - Porting applications to Distributed Computing Infrastructures (DCIs) is eased by the use of workflow abstractions. Yet, estimating the impact of the execution DCI on application performance is difficult due to the heterogeneity of the resources available, middleware and operation models. This paper describes a workflow-based experimental method to acquire objective performance comparison criterions when dealing with completely different DCIs. Experiments were conducted on the European EGI and the French Grid'5000 infrastructures to highlight raw performance variations and identify their causes. The results obtained also show that it is possible to conduct experiments on a production infrastructure with similar reproducibility as on an experimental platform.

KW - C.4 [computer systems organization]

KW - Distributed computing infrastructures

KW - Performance

KW - Performance of systems

KW - Workflows

UR - http://www.scopus.com/inward/record.url?scp=78751537733&partnerID=8YFLogxK

U2 - 10.1109/WORKS.2010.5671856

DO - 10.1109/WORKS.2010.5671856

M3 - Conference contribution

AN - SCOPUS:78751537733

SN - 9781424489893

T3 - 2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010

BT - 2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010

T2 - 2010 5th Workshop on Workflows in Support of Large-Scale Science, WORKS 2010

Y2 - 14 November 2010 through 14 November 2010

ER -

Workflow-based comparison of two distributed computing infrastructures

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this