-
H. Casanova,
G. Obertelli,
F. Berman,
and R. Wolski.
The AppLeS Parameter Sweep Template: User-Level Middleware for the Grid.
Scientific Programming,
8(3):111--126,
2000.
Keywords:
grid middleware,
application level scheduling,
adaptive runtime scheduling,
computational grid,
grid middleware,
parameter sweeps application,
scheduling heuristics,
distributed storage,
best paper award,.
| Abstract: |
The Computational Grid is a promising platform for the efficient execution of parameter sweep applications over large parameter spaces. To achieve performance on the Grid, such applications must be scheduled so that shared data files are strategically placed to maximize re-use, and so that the application execution can adapt to the deliverable performance potential of target heterogeneous, distributed and shared resources. Parameter sweep applications are an important class of applications and would greatly benefit from the development of Grid middleware that embeds a scheduler for performance and targets Grid resources transparently.\par In this paper we describe a user-level Grid middleware project, the AppLeS Parameter Sweep Template (APST), that uses application-level scheduling techniques and various Grid technologies to allow the efficient deployment of parameter sweep applications over the Grid. We discuss several possible scheduling algorithms and detail our software design. We then describe our current implementation of APST using systems like Globus, NetSolve and the Network Weather Service, and present experimental results.\par Best paper award, reproduced and slightly upgraded from the Supercomputing'2000 Proceedings (CD-ROM), IEEE and ACM SIGARCH, Dallas, TX, November 2000. |
@Article{ ArticleCasanova.COBW_SP00,
abstract = {The Computational Grid is a promising platform for the efficient execution of parameter sweep applications over large parameter spaces. To achieve performance on the Grid, such applications must be scheduled so that shared data files are strategically placed to maximize re-use, and so that the application execution can adapt to the deliverable performance potential of target heterogeneous, distributed and shared resources. Parameter sweep applications are an important class of applications and would greatly benefit from the development of Grid middleware that embeds a scheduler for performance and targets Grid resources transparently.\par In this paper we describe a user-level Grid middleware project, the AppLeS Parameter Sweep Template (APST), that uses application-level scheduling techniques and various Grid technologies to allow the efficient deployment of parameter sweep applications over the Grid. We discuss several possible scheduling algorithms and detail our software design. We then describe our current implementation of APST using systems like Globus, NetSolve and the Network Weather Service, and present experimental results.\par Best paper award, reproduced and slightly upgraded from the Supercomputing'2000 Proceedings (CD-ROM), IEEE and ACM SIGARCH, Dallas, TX, November 2000.},
author = {H. Casanova and G. Obertelli and F. Berman and R. Wolski},
journal = {Scientific Programming},
keywords = {grid middleware, application level scheduling, adaptive runtime scheduling, computational grid, grid middleware, parameter sweeps application, scheduling heuristics, distributed storage, best paper award,},
number = {3},
pages = {111--126},
title = {The AppLe{S} Parameter Sweep Template: User-Level Middleware for the Grid},
volume = {8},
year = {2000}
}
-
A. Chervenak,
I. Foster,
C. Kesselman,
C. Salisbury,
and S. Tuecke.
The Data Grid: Towards an architecture for the distributed management and analysis of large scientific datasets.
Journal of Network and Computer Applications,
23(3):187--200,
2000.
@Article{ ArticleChervenak.CFKST_JNCA00,
author = {Chervenak, A. and Foster, I. and Kesselman, C. and Salisbury, C. and Tuecke, S.},
journal = {Journal of Network and Computer Applications},
number = {3},
pages = {187--200},
publisher = {Elsevier},
title = {The Data Grid: Towards an architecture for the distributed management and analysis of large scientific datasets},
volume = {23},
year = {2000}
}
-
Eddy Caron,
Dominique Lazure,
and Frédéric Suter.
Manipulation de données de grande taille dans Scilab//.
In Douzièmes Rencontres Francophones du Parallélisme,
Besançon,
June 2000.
Keywords:
Scilab//.
@InProceedings{ InProceedingsCaron.CLS_00,
address = {Besan{\c c}on},
author = {Caron, Eddy and Lazure, Dominique and Suter, Fr{\'e}d{\'e}ric},
booktitle = {Douzi{\`e}mes Rencontres Francophones du Parall{\'e}lisme},
keywords = {Scilab//},
month = jun,
title = {Manipulation de donn{\'e}es de grande taille dans Scilab//},
year = 2000
}
-
Eddy Caron,
Dominique Lazure,
and Gil Utard.
Inversion of Huge Matrix on Cluster.
In Cluster 2000. IEEE International Conference on Cluster Computing. Technische Universität Chemnitz, Saxony, Germany. (Poster),
November 2000.
Keywords:
Out-of-core.
@InProceedings{ InProceedingsCaron.CLU_00*1,
author = {Caron, Eddy and Lazure, Dominique and Utard, Gil},
booktitle = {{Cluster 2000}. IEEE International Conference on Cluster Computing. Technische Universit{\"a}t Chemnitz, Saxony, Germany. (Poster)},
keywords = {Out-of-core},
month = nov,
title = {Inversion of Huge Matrix on Cluster},
year = 2000
}
-
Eddy Caron,
Dominique Lazure,
and Gil Utard.
Performance Modeling and Analysis of Parallel Out-of-Core Matrix Factorization.
In HiPC'2000. 7th International Conference on High Performance Computing. Bangalore, India,
December 2000.
Keywords:
Out-of-core.
@InProceedings{ InProceedingsCaron.CLU_00*2,
author = {Caron, Eddy and Lazure, Dominique and Utard, Gil},
booktitle = {{HiPC'2000}. 7th International Conference on High Performance Computing. Bangalore, India},
keywords = {Out-of-core},
month = dec,
postscript = {http://www.ens-lyon.fr/~ecaron/biblio/papers/hipc00.ps.gz},
title = {Performance Modeling and Analysis of Parallel Out-of-Core Matrix Factorization},
year = {2000}
}
-
Eddy Caron,
Dominique Lazure,
and Gil Utard.
Modélisation et analyse des performances de la factorisation parallel LU out-of-core.
In RenPar 2000. 12èmes Rencontres francophones du parallélisme. Besançon. France Rencontres francophones du parallélisme.,
Besançon. France,
June 2000.
Keywords:
Out-of-core.
@InProceedings{ InProceedingsCaron.CLU_00,
address = {Besan{\c c}on. France},
author = {Caron, Eddy and Lazure, Dominique and Utard, Gil},
booktitle = {RenPar 2000. 12{\`e}mes Rencontres francophones du parall{\'e}lisme. Besan{\c c}on. France Rencontres francophones du parall{\'e}lisme.},
keywords = {Out-of-core},
month = jun,
title = {Mod{\'e}lisation et analyse des performances de la factorisation parallel LU out-of-core},
year = {2000}
}
-
Henri Casanova,
Arnaud Legrand,
Dmitrii Zagorodnov,
and Francine Berman.
Heuristics for Scheduling Parameter Sweep Applications in Grid Environments.
In Heterogeneous Computing Workshop,
pages 349--363,
2000.
| Abstract: |
The Computational Grid provides a promising platform for the efficient execution of parameter sweep applications over very large parameter spaces. Scheduling such applications is challenging because target resources are heterogeneous, because their load and availability varies dynamically, and because independent tasks may share common data files. In this paper, we propose an adaptive scheduling algorithm for parameter sweep applications on the Grid. We modify standard heuristics for task/host assignment in perfectly predictable environments (Max-min, Min-min, Sufferage), and we propose an extension of Sufferage called XSufferage. Using simulation, we demonstrate that XSufferage can take advantage of file sharing to achieve better performance than the other heuristics. We also study the impact of inaccurate performance prediction on scheduling. Our study shows that: (i) different heuristics behave differently when predictions are inaccurate; (ii) increased adaptivity leads to better performance. |
@InProceedings{ InProceedingsCasanova.CLZB_00,
abstract = {The Computational Grid provides a promising platform for the efficient execution of parameter sweep applications over very large parameter spaces. Scheduling such applications is challenging because target resources are heterogeneous, because their load and availability varies dynamically, and because independent tasks may share common data files. In this paper, we propose an adaptive scheduling algorithm for parameter sweep applications on the Grid. We modify standard heuristics for task/host assignment in perfectly predictable environments (Max-min, Min-min, Sufferage), and we propose an extension of Sufferage called XSufferage. Using simulation, we demonstrate that XSufferage can take advantage of file sharing to achieve better performance than the other heuristics. We also study the impact of inaccurate performance prediction on scheduling. Our study shows that: (i) different heuristics behave differently when predictions are inaccurate; (ii) increased adaptivity leads to better performance.},
author = {Henri Casanova and Arnaud Legrand and Dmitrii Zagorodnov and Francine Berman},
booktitle = {Heterogeneous Computing Workshop},
pages = {349--363},
postscript = {http://apples.ucsd.edu/pubs/hcw00_pst.ps},
title = {Heuristics for Scheduling Parameter Sweep Applications in Grid Environments},
year = {2000}
}
-
N. Krasnogor and J. Smith.
A Java Memetic Algorithms Framework.
In Workshop Program, Proceedings of the 2000 Genetic and Evolutionary Computation Conference. Morgan Kaufmann,
2000.
@Conference{ Con-Krasnogor.KS_00,
author = {Krasnogor, N. and Smith, J.},
booktitle = {Workshop Program, Proceedings of the 2000 Genetic and Evolutionary Computation Conference. Morgan Kaufmann},
title = {A Java Memetic Algorithms Framework},
year = {2000}
}