The design and implementation of the KOALA co-allocating grid scheduler

H.H. Mohamed; D.H.J. Epema

doi:10.1007/11508380_65

The design and implementation of the KOALA co-allocating grid scheduler

H.H. Mohamed, D.H.J. Epema

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

23 Citations (Scopus)

8 Downloads (Pure)

Abstract

In multicluster systems, and more generally, in grids, jobs may require co-allocation, i.e., the simultaneous allocation of resources such as processors and input .les in multiple clusters. While such jobs may have reduced runtimes because they have access to more resources, waiting for processors in multiple clusters and for the input .les to become available in the right locations, may introduce ine.ciencies. In this paper we present the design of KOALA, a prototype for processor and data co-allocation that tries to minimize these ine.ciencies through the use of its Close-to-Files placement policy and its Incremental Claiming Policy. The latter policy tries to solve the problem of a lack of support for reservation by local resource managers.

Original language	English
Title of host publication	Advances in Grid Computing - EGC 2005 (European Grid Conference, Amsterdam, The Netherlands, February 14-16, 2005. Revised Selected Papers)
Editors	P.M.A. Sloot, A.G. Hoekstra, T. Priol, A. Reinefeld, M. Bubak
Place of Publication	Berlin
Publisher	Springer
Pages	640-650
ISBN (Print)	3-540-26918-5
DOIs	https://doi.org/10.1007/11508380_65
Publication status	Published - 2005

Publication series

Name	Lecture Notes in Computer Science
Volume	3470
ISSN (Print)	0302-9743

Access to Document

10.1007/11508380_65

Cite this

Mohamed, H. H., & Epema, D. H. J. (2005). The design and implementation of the KOALA co-allocating grid scheduler. In P. M. A. Sloot, A. G. Hoekstra, T. Priol, A. Reinefeld, & M. Bubak (Eds.), Advances in Grid Computing - EGC 2005 (European Grid Conference, Amsterdam, The Netherlands, February 14-16, 2005. Revised Selected Papers) (pp. 640-650). (Lecture Notes in Computer Science; Vol. 3470). Springer. https://doi.org/10.1007/11508380_65

Mohamed, H.H. ; Epema, D.H.J. / The design and implementation of the KOALA co-allocating grid scheduler. Advances in Grid Computing - EGC 2005 (European Grid Conference, Amsterdam, The Netherlands, February 14-16, 2005. Revised Selected Papers). editor / P.M.A. Sloot ; A.G. Hoekstra ; T. Priol ; A. Reinefeld ; M. Bubak. Berlin : Springer, 2005. pp. 640-650 (Lecture Notes in Computer Science).

@inproceedings{cad0838f9f3d4b9587ca2f49b4102891,

title = "The design and implementation of the KOALA co-allocating grid scheduler",

abstract = "In multicluster systems, and more generally, in grids, jobs may require co-allocation, i.e., the simultaneous allocation of resources such as processors and input .les in multiple clusters. While such jobs may have reduced runtimes because they have access to more resources, waiting for processors in multiple clusters and for the input .les to become available in the right locations, may introduce ine.ciencies. In this paper we present the design of KOALA, a prototype for processor and data co-allocation that tries to minimize these ine.ciencies through the use of its Close-to-Files placement policy and its Incremental Claiming Policy. The latter policy tries to solve the problem of a lack of support for reservation by local resource managers.",

author = "H.H. Mohamed and D.H.J. Epema",

year = "2005",

doi = "10.1007/11508380_65",

language = "English",

isbn = "3-540-26918-5",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "640--650",

editor = "P.M.A. Sloot and A.G. Hoekstra and T. Priol and A. Reinefeld and M. Bubak",

booktitle = "Advances in Grid Computing - EGC 2005 (European Grid Conference, Amsterdam, The Netherlands, February 14-16, 2005. Revised Selected Papers)",

address = "Germany",

}

Mohamed, HH & Epema, DHJ 2005, The design and implementation of the KOALA co-allocating grid scheduler. in PMA Sloot, AG Hoekstra, T Priol, A Reinefeld & M Bubak (eds), Advances in Grid Computing - EGC 2005 (European Grid Conference, Amsterdam, The Netherlands, February 14-16, 2005. Revised Selected Papers). Lecture Notes in Computer Science, vol. 3470, Springer, Berlin, pp. 640-650. https://doi.org/10.1007/11508380_65

The design and implementation of the KOALA co-allocating grid scheduler. / Mohamed, H.H.; Epema, D.H.J.
Advances in Grid Computing - EGC 2005 (European Grid Conference, Amsterdam, The Netherlands, February 14-16, 2005. Revised Selected Papers). ed. / P.M.A. Sloot; A.G. Hoekstra; T. Priol; A. Reinefeld; M. Bubak. Berlin: Springer, 2005. p. 640-650 (Lecture Notes in Computer Science; Vol. 3470).

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

TY - GEN

T1 - The design and implementation of the KOALA co-allocating grid scheduler

AU - Mohamed, H.H.

AU - Epema, D.H.J.

PY - 2005

Y1 - 2005

N2 - In multicluster systems, and more generally, in grids, jobs may require co-allocation, i.e., the simultaneous allocation of resources such as processors and input .les in multiple clusters. While such jobs may have reduced runtimes because they have access to more resources, waiting for processors in multiple clusters and for the input .les to become available in the right locations, may introduce ine.ciencies. In this paper we present the design of KOALA, a prototype for processor and data co-allocation that tries to minimize these ine.ciencies through the use of its Close-to-Files placement policy and its Incremental Claiming Policy. The latter policy tries to solve the problem of a lack of support for reservation by local resource managers.

AB - In multicluster systems, and more generally, in grids, jobs may require co-allocation, i.e., the simultaneous allocation of resources such as processors and input .les in multiple clusters. While such jobs may have reduced runtimes because they have access to more resources, waiting for processors in multiple clusters and for the input .les to become available in the right locations, may introduce ine.ciencies. In this paper we present the design of KOALA, a prototype for processor and data co-allocation that tries to minimize these ine.ciencies through the use of its Close-to-Files placement policy and its Incremental Claiming Policy. The latter policy tries to solve the problem of a lack of support for reservation by local resource managers.

U2 - 10.1007/11508380_65

DO - 10.1007/11508380_65

M3 - Conference contribution

SN - 3-540-26918-5

T3 - Lecture Notes in Computer Science

SP - 640

EP - 650

BT - Advances in Grid Computing - EGC 2005 (European Grid Conference, Amsterdam, The Netherlands, February 14-16, 2005. Revised Selected Papers)

A2 - Sloot, P.M.A.

A2 - Hoekstra, A.G.

A2 - Priol, T.

A2 - Reinefeld, A.

A2 - Bubak, M.

PB - Springer

CY - Berlin

ER -

Mohamed HH, Epema DHJ. The design and implementation of the KOALA co-allocating grid scheduler. In Sloot PMA, Hoekstra AG, Priol T, Reinefeld A, Bubak M, editors, Advances in Grid Computing - EGC 2005 (European Grid Conference, Amsterdam, The Netherlands, February 14-16, 2005. Revised Selected Papers). Berlin: Springer. 2005. p. 640-650. (Lecture Notes in Computer Science). doi: 10.1007/11508380_65

The design and implementation of the KOALA co-allocating grid scheduler

Abstract

Publication series

Access to Document

Fingerprint

Cite this