Scheduling jobs in the cloud using on-demand and reserved instances

S. Shen, K. Deng, A. Iosup, D.H.J. Epema

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

26 Citations (Scopus)

Abstract

Deploying applications in leased cloud infrastructure is increasingly considered by a variety of business and service integrators. However, the challenge of selecting the leasing strategy — larger or faster instances? on-demand or reserved instances? etc.— and to configure the leasing strategy with appropriate scheduling policies is still daunting for the (potential) cloud user. In this work, we investigate leasing strategies and their policies from a broker’s perspective. We propose, CoH, a family of Cloud-based, online, Hybrid scheduling policies that minimizes rental cost by making use of both on-demand and reserved instances. We formulate the resource provisioning and job allocation policies as Integer Programming problems. As the policies need to be executed online, we limit the time to explore the optimal solution of the integer program, and compare the obtained solution with various heuristics-based policies; then automatically pick the best one. We show, via simulation and using multiple real-world traces, that the hybrid leasing policy can obtain significantly lower cost than typical heuristics-based policies.
Original languageEnglish
Title of host publicationEuro-Par 2013 Parallel Processing (19th International Conference, Aachen, Germany, August 26-30, 2013. Proceedings)
EditorsF. Wolf, B. Mohr, D. Mey, an
Place of PublicationBerlin
PublisherSpringer
Pages242-254
ISBN (Print)978-3-642-40046-9
DOIs
Publication statusPublished - 2013

Publication series

NameLecture Notes in Computer Science
Volume8097
ISSN (Print)0302-9743

Fingerprint

Scheduling
Integer programming
Costs
Industry

Cite this

Shen, S., Deng, K., Iosup, A., & Epema, D. H. J. (2013). Scheduling jobs in the cloud using on-demand and reserved instances. In F. Wolf, B. Mohr, & D. Mey, an (Eds.), Euro-Par 2013 Parallel Processing (19th International Conference, Aachen, Germany, August 26-30, 2013. Proceedings) (pp. 242-254). (Lecture Notes in Computer Science; Vol. 8097). Berlin: Springer. https://doi.org/10.1007/978-3-642-40047-6_27
Shen, S. ; Deng, K. ; Iosup, A. ; Epema, D.H.J. / Scheduling jobs in the cloud using on-demand and reserved instances. Euro-Par 2013 Parallel Processing (19th International Conference, Aachen, Germany, August 26-30, 2013. Proceedings). editor / F. Wolf ; B. Mohr ; D. Mey, an. Berlin : Springer, 2013. pp. 242-254 (Lecture Notes in Computer Science).
@inproceedings{200036dfe5ac49cabfda2b0ceed529fc,
title = "Scheduling jobs in the cloud using on-demand and reserved instances",
abstract = "Deploying applications in leased cloud infrastructure is increasingly considered by a variety of business and service integrators. However, the challenge of selecting the leasing strategy — larger or faster instances? on-demand or reserved instances? etc.— and to configure the leasing strategy with appropriate scheduling policies is still daunting for the (potential) cloud user. In this work, we investigate leasing strategies and their policies from a broker’s perspective. We propose, CoH, a family of Cloud-based, online, Hybrid scheduling policies that minimizes rental cost by making use of both on-demand and reserved instances. We formulate the resource provisioning and job allocation policies as Integer Programming problems. As the policies need to be executed online, we limit the time to explore the optimal solution of the integer program, and compare the obtained solution with various heuristics-based policies; then automatically pick the best one. We show, via simulation and using multiple real-world traces, that the hybrid leasing policy can obtain significantly lower cost than typical heuristics-based policies.",
author = "S. Shen and K. Deng and A. Iosup and D.H.J. Epema",
year = "2013",
doi = "10.1007/978-3-642-40047-6_27",
language = "English",
isbn = "978-3-642-40046-9",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "242--254",
editor = "F. Wolf and B. Mohr and {Mey, an}, D.",
booktitle = "Euro-Par 2013 Parallel Processing (19th International Conference, Aachen, Germany, August 26-30, 2013. Proceedings)",
address = "Germany",

}

Shen, S, Deng, K, Iosup, A & Epema, DHJ 2013, Scheduling jobs in the cloud using on-demand and reserved instances. in F Wolf, B Mohr & D Mey, an (eds), Euro-Par 2013 Parallel Processing (19th International Conference, Aachen, Germany, August 26-30, 2013. Proceedings). Lecture Notes in Computer Science, vol. 8097, Springer, Berlin, pp. 242-254. https://doi.org/10.1007/978-3-642-40047-6_27

Scheduling jobs in the cloud using on-demand and reserved instances. / Shen, S.; Deng, K.; Iosup, A.; Epema, D.H.J.

Euro-Par 2013 Parallel Processing (19th International Conference, Aachen, Germany, August 26-30, 2013. Proceedings). ed. / F. Wolf; B. Mohr; D. Mey, an. Berlin : Springer, 2013. p. 242-254 (Lecture Notes in Computer Science; Vol. 8097).

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Scheduling jobs in the cloud using on-demand and reserved instances

AU - Shen, S.

AU - Deng, K.

AU - Iosup, A.

AU - Epema, D.H.J.

PY - 2013

Y1 - 2013

N2 - Deploying applications in leased cloud infrastructure is increasingly considered by a variety of business and service integrators. However, the challenge of selecting the leasing strategy — larger or faster instances? on-demand or reserved instances? etc.— and to configure the leasing strategy with appropriate scheduling policies is still daunting for the (potential) cloud user. In this work, we investigate leasing strategies and their policies from a broker’s perspective. We propose, CoH, a family of Cloud-based, online, Hybrid scheduling policies that minimizes rental cost by making use of both on-demand and reserved instances. We formulate the resource provisioning and job allocation policies as Integer Programming problems. As the policies need to be executed online, we limit the time to explore the optimal solution of the integer program, and compare the obtained solution with various heuristics-based policies; then automatically pick the best one. We show, via simulation and using multiple real-world traces, that the hybrid leasing policy can obtain significantly lower cost than typical heuristics-based policies.

AB - Deploying applications in leased cloud infrastructure is increasingly considered by a variety of business and service integrators. However, the challenge of selecting the leasing strategy — larger or faster instances? on-demand or reserved instances? etc.— and to configure the leasing strategy with appropriate scheduling policies is still daunting for the (potential) cloud user. In this work, we investigate leasing strategies and their policies from a broker’s perspective. We propose, CoH, a family of Cloud-based, online, Hybrid scheduling policies that minimizes rental cost by making use of both on-demand and reserved instances. We formulate the resource provisioning and job allocation policies as Integer Programming problems. As the policies need to be executed online, we limit the time to explore the optimal solution of the integer program, and compare the obtained solution with various heuristics-based policies; then automatically pick the best one. We show, via simulation and using multiple real-world traces, that the hybrid leasing policy can obtain significantly lower cost than typical heuristics-based policies.

U2 - 10.1007/978-3-642-40047-6_27

DO - 10.1007/978-3-642-40047-6_27

M3 - Conference contribution

SN - 978-3-642-40046-9

T3 - Lecture Notes in Computer Science

SP - 242

EP - 254

BT - Euro-Par 2013 Parallel Processing (19th International Conference, Aachen, Germany, August 26-30, 2013. Proceedings)

A2 - Wolf, F.

A2 - Mohr, B.

A2 - Mey, an, D.

PB - Springer

CY - Berlin

ER -

Shen S, Deng K, Iosup A, Epema DHJ. Scheduling jobs in the cloud using on-demand and reserved instances. In Wolf F, Mohr B, Mey, an D, editors, Euro-Par 2013 Parallel Processing (19th International Conference, Aachen, Germany, August 26-30, 2013. Proceedings). Berlin: Springer. 2013. p. 242-254. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-642-40047-6_27