Toward autonomic grids: Analyzing the job flow with affinity streaming

Xiangliang Zhang; Cyril Furtlehner; Julien Perez; Cecile Germain-Renaud; Michèle Sebag

doi:10.1145/1557019.1557126

Toward autonomic grids: Analyzing the job flow with affinity streaming

Xiangliang Zhang^*, Cyril Furtlehner, Julien Perez, Cecile Germain-Renaud, Michèle Sebag

^*Corresponding author for this work

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

21 Scopus citations

Abstract

The Afinity Propagation (AP) clustering algorithm proposed by Frey and Dueck (2007) provides an understandable, nearly optimal summary of a dataset, albeit with quadratic compu- tational complexity. This paper, motivated by Autonomic Computing, extends AP to the data streaming framework. Firstly a hierarchical strategy is used to reduce the complex- ity to O(N ^1+ε); the distortion loss incurred is analyzed in relation with the dimension of the data items. Secondly, a coupling with a change detection test is used to cope with non-stationary data distribution, and rebuild the model as needed. The presented approach Strap is applied to the stream of jobs submitted to the EGEE Grid, providing an understandable description of the job ow and enabling the system administrator to spot online some sources of fail- ures.

Original language	English (US)
Title of host publication	KDD '09
Subtitle of host publication	Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Pages	987-995
Number of pages	9
DOIs	https://doi.org/10.1145/1557019.1557126
State	Published - 2009
Externally published	Yes
Event	15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '09 - Paris, France Duration: Jun 28 2009 → Jul 1 2009

Publication series

Name	Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Other

Other	15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '09
Country/Territory	France
City	Paris
Period	06/28/09 → 07/1/09

Keywords

Afinity propagation
Autonomic computing
Online clus- tering

ASJC Scopus subject areas

Software
Information Systems

Access to Document

10.1145/1557019.1557126

Cite this

Zhang, X., Furtlehner, C., Perez, J., Germain-Renaud, C., & Sebag, M. (2009). Toward autonomic grids: Analyzing the job flow with affinity streaming. In KDD '09: Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 987-995). (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). https://doi.org/10.1145/1557019.1557126

@inproceedings{7baa1175e6104a88a0850c00ebff9d02,

title = "Toward autonomic grids: Analyzing the job flow with affinity streaming",

abstract = "The Afinity Propagation (AP) clustering algorithm proposed by Frey and Dueck (2007) provides an understandable, nearly optimal summary of a dataset, albeit with quadratic compu- tational complexity. This paper, motivated by Autonomic Computing, extends AP to the data streaming framework. Firstly a hierarchical strategy is used to reduce the complex- ity to O(N 1+ε); the distortion loss incurred is analyzed in relation with the dimension of the data items. Secondly, a coupling with a change detection test is used to cope with non-stationary data distribution, and rebuild the model as needed. The presented approach Strap is applied to the stream of jobs submitted to the EGEE Grid, providing an understandable description of the job ow and enabling the system administrator to spot online some sources of fail- ures.",

keywords = "Afinity propagation, Autonomic computing, Online clus- tering",

author = "Xiangliang Zhang and Cyril Furtlehner and Julien Perez and Cecile Germain-Renaud and Mich{\`e}le Sebag",

year = "2009",

doi = "10.1145/1557019.1557126",

language = "English (US)",

isbn = "9781605584959",

series = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

pages = "987--995",

booktitle = "KDD '09",

note = "15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '09 ; Conference date: 28-06-2009 Through 01-07-2009",

}

Zhang, X, Furtlehner, C, Perez, J, Germain-Renaud, C & Sebag, M 2009, Toward autonomic grids: Analyzing the job flow with affinity streaming. in KDD '09: Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 987-995, 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '09, Paris, France, 06/28/09. https://doi.org/10.1145/1557019.1557126

Toward autonomic grids: Analyzing the job flow with affinity streaming. / Zhang, Xiangliang; Furtlehner, Cyril; Perez, Julien et al.
KDD '09: Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2009. p. 987-995 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

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

TY - GEN

T1 - Toward autonomic grids

T2 - 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '09

AU - Zhang, Xiangliang

AU - Furtlehner, Cyril

AU - Perez, Julien

AU - Germain-Renaud, Cecile

AU - Sebag, Michèle

PY - 2009

Y1 - 2009

N2 - The Afinity Propagation (AP) clustering algorithm proposed by Frey and Dueck (2007) provides an understandable, nearly optimal summary of a dataset, albeit with quadratic compu- tational complexity. This paper, motivated by Autonomic Computing, extends AP to the data streaming framework. Firstly a hierarchical strategy is used to reduce the complex- ity to O(N 1+ε); the distortion loss incurred is analyzed in relation with the dimension of the data items. Secondly, a coupling with a change detection test is used to cope with non-stationary data distribution, and rebuild the model as needed. The presented approach Strap is applied to the stream of jobs submitted to the EGEE Grid, providing an understandable description of the job ow and enabling the system administrator to spot online some sources of fail- ures.

AB - The Afinity Propagation (AP) clustering algorithm proposed by Frey and Dueck (2007) provides an understandable, nearly optimal summary of a dataset, albeit with quadratic compu- tational complexity. This paper, motivated by Autonomic Computing, extends AP to the data streaming framework. Firstly a hierarchical strategy is used to reduce the complex- ity to O(N 1+ε); the distortion loss incurred is analyzed in relation with the dimension of the data items. Secondly, a coupling with a change detection test is used to cope with non-stationary data distribution, and rebuild the model as needed. The presented approach Strap is applied to the stream of jobs submitted to the EGEE Grid, providing an understandable description of the job ow and enabling the system administrator to spot online some sources of fail- ures.

KW - Afinity propagation

KW - Autonomic computing

KW - Online clus- tering

UR - http://www.scopus.com/inward/record.url?scp=71049172937&partnerID=8YFLogxK

U2 - 10.1145/1557019.1557126

DO - 10.1145/1557019.1557126

M3 - Conference contribution

AN - SCOPUS:71049172937

SN - 9781605584959

T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

SP - 987

EP - 995

BT - KDD '09

Y2 - 28 June 2009 through 1 July 2009

ER -

Zhang X, Furtlehner C, Perez J, Germain-Renaud C, Sebag M. Toward autonomic grids: Analyzing the job flow with affinity streaming. In KDD '09: Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2009. p. 987-995. (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). doi: 10.1145/1557019.1557126

Toward autonomic grids: Analyzing the job flow with affinity streaming

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this