A hypergraph partitioning based approach for scheduling of tasks with batch-shared I/O

Gaurav Khanna; Nagavijayalakshmi Vydyanathan; Tahsin Kurc; Umit Catalyurek; Pete Wyckoff; Joel Saltz; P. Sadayappan

doi:10.1109/CCGRID.2005.1558643

A hypergraph partitioning based approach for scheduling of tasks with batch-shared I/O

Gaurav Khanna
, Nagavijayalakshmi Vydyanathan
, Tahsin Kurc
, Umit Catalyurek
, Pete Wyckoff
, Joel Saltz
, P. Sadayappan

Biomedical Informatics

Ohio State University
Ohio Supercomputer Center

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

27 Scopus citations

Abstract

This paper proposes a novel, hypergraph partitioning based strategy to schedule multiple data analysis tasks with batch-shared I/O behavior. This strategy formulates the sharing of files among tasks as a hypergraph to minimize the I/O overheads due to transferring of the same set of files multiple times and employs a dynamic scheme for file transfers to reduce contention on the storage system. We experimentally evaluate the proposed approach using application emulators from two application domains; analysis of remotely-sensed data and biomedical imaging.

Original language	English
Title of host publication	2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005
Pages	792-799
Number of pages	8
DOIs	https://doi.org/10.1109/CCGRID.2005.1558643
State	Published - 2005
Event	2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005 - Cardiff, Wales, United Kingdom Duration: May 9 2005 → May 12 2005

Publication series

Name	2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005
Volume	2

Conference

Conference	2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005
Country/Territory	United Kingdom
City	Cardiff, Wales
Period	05/9/05 → 05/12/05

Access to Document

10.1109/CCGRID.2005.1558643

Cite this

Khanna, G., Vydyanathan, N., Kurc, T., Catalyurek, U., Wyckoff, P., Saltz, J., & Sadayappan, P. (2005). A hypergraph partitioning based approach for scheduling of tasks with batch-shared I/O. In 2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005 (pp. 792-799). Article 1558643 (2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005; Vol. 2). https://doi.org/10.1109/CCGRID.2005.1558643

@inproceedings{56961307ecb24372b9bfdb724c123b1b,

title = "A hypergraph partitioning based approach for scheduling of tasks with batch-shared I/O",

abstract = "This paper proposes a novel, hypergraph partitioning based strategy to schedule multiple data analysis tasks with batch-shared I/O behavior. This strategy formulates the sharing of files among tasks as a hypergraph to minimize the I/O overheads due to transferring of the same set of files multiple times and employs a dynamic scheme for file transfers to reduce contention on the storage system. We experimentally evaluate the proposed approach using application emulators from two application domains; analysis of remotely-sensed data and biomedical imaging.",

author = "Gaurav Khanna and Nagavijayalakshmi Vydyanathan and Tahsin Kurc and Umit Catalyurek and Pete Wyckoff and Joel Saltz and P. Sadayappan",

year = "2005",

doi = "10.1109/CCGRID.2005.1558643",

language = "English",

isbn = "0780390741",

series = "2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005",

pages = "792--799",

booktitle = "2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005",

note = "2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005 ; Conference date: 09-05-2005 Through 12-05-2005",

}

Khanna, G, Vydyanathan, N, Kurc, T, Catalyurek, U, Wyckoff, P, Saltz, J & Sadayappan, P 2005, A hypergraph partitioning based approach for scheduling of tasks with batch-shared I/O. in 2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005., 1558643, 2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005, vol. 2, pp. 792-799, 2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005, Cardiff, Wales, United Kingdom, 05/9/05. https://doi.org/10.1109/CCGRID.2005.1558643

A hypergraph partitioning based approach for scheduling of tasks with batch-shared I/O. / Khanna, Gaurav; Vydyanathan, Nagavijayalakshmi; Kurc, Tahsin et al.
2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005. 2005. p. 792-799 1558643 (2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005; Vol. 2).

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

TY - GEN

T1 - A hypergraph partitioning based approach for scheduling of tasks with batch-shared I/O

AU - Khanna, Gaurav

AU - Vydyanathan, Nagavijayalakshmi

AU - Kurc, Tahsin

AU - Catalyurek, Umit

AU - Wyckoff, Pete

AU - Saltz, Joel

AU - Sadayappan, P.

PY - 2005

Y1 - 2005

N2 - This paper proposes a novel, hypergraph partitioning based strategy to schedule multiple data analysis tasks with batch-shared I/O behavior. This strategy formulates the sharing of files among tasks as a hypergraph to minimize the I/O overheads due to transferring of the same set of files multiple times and employs a dynamic scheme for file transfers to reduce contention on the storage system. We experimentally evaluate the proposed approach using application emulators from two application domains; analysis of remotely-sensed data and biomedical imaging.

AB - This paper proposes a novel, hypergraph partitioning based strategy to schedule multiple data analysis tasks with batch-shared I/O behavior. This strategy formulates the sharing of files among tasks as a hypergraph to minimize the I/O overheads due to transferring of the same set of files multiple times and employs a dynamic scheme for file transfers to reduce contention on the storage system. We experimentally evaluate the proposed approach using application emulators from two application domains; analysis of remotely-sensed data and biomedical imaging.

UR - https://www.scopus.com/pages/publications/33845325178

U2 - 10.1109/CCGRID.2005.1558643

DO - 10.1109/CCGRID.2005.1558643

M3 - Conference contribution

AN - SCOPUS:33845325178

SN - 0780390741

SN - 9780780390744

T3 - 2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005

SP - 792

EP - 799

BT - 2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005

T2 - 2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005

Y2 - 9 May 2005 through 12 May 2005

ER -

Khanna G, Vydyanathan N, Kurc T, Catalyurek U, Wyckoff P, Saltz J et al. A hypergraph partitioning based approach for scheduling of tasks with batch-shared I/O. In 2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005. 2005. p. 792-799. 1558643. (2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005). doi: 10.1109/CCGRID.2005.1558643

A hypergraph partitioning based approach for scheduling of tasks with batch-shared I/O

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this