Auction-based cloud service differentiation with service level objectives

Jianbing Ding, Zhenjie Zhang, T. B Ma Richard, Yin Yang

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

The emergence of the cloud computing paradigm has greatly enabled innovative service models, such as Platform as a Service (PaaS), and distributed computing frameworks, such as MapReduce. However, most existing cloud systems fail to distinguish users with different preferences, or jobs of different natures. Consequently, they are unable to provide service differentiation, leading to inefficient allocations of cloud resources. Moreover, contentions on the resources exacerbate this inefficiency, when prioritizing crucial jobs is necessary, but impossible. Motivated by this, we propose Abacus, a generic resource management framework addressing this problem. Abacus interacts with users through an auction mechanism, which allows users to specify their priorities using budgets, and job characteristics via utility functions. Based on this information, Abacus computes the optimal allocation and scheduling of resources. Meanwhile, the auction mechanism in Abacus possesses important properties including incentive compatibility (i.e., the users' best strategy is to simply bid their true budgets and job utilities) and monotonicity (i.e., users are motivated to increase their budgets in order to receive better services). In addition, when the user is unclear about her utility function, Abacus automatically learns this function based on statistics of her previous jobs. Extensive experiments, running Hadoop on a private cluster and Amazon EC2, demonstrate the high performance and other desirable properties of Abacus.

Original languageEnglish
Pages (from-to)231-249
Number of pages19
JournalComputer Networks
Volume94
DOIs
Publication statusPublished - 15 Jan 2016
Externally publishedYes

Fingerprint

Distributed computer systems
Cloud computing
Scheduling
Statistics
Experiments

Keywords

  • Auction
  • Cloud computing
  • MapReduce
  • Service differentiation

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Auction-based cloud service differentiation with service level objectives. / Ding, Jianbing; Zhang, Zhenjie; Richard, T. B Ma; Yang, Yin.

In: Computer Networks, Vol. 94, 15.01.2016, p. 231-249.

Research output: Contribution to journalArticle

Ding, Jianbing ; Zhang, Zhenjie ; Richard, T. B Ma ; Yang, Yin. / Auction-based cloud service differentiation with service level objectives. In: Computer Networks. 2016 ; Vol. 94. pp. 231-249.
@article{ae484e925f19445b9ee83f1bcdef5bde,
title = "Auction-based cloud service differentiation with service level objectives",
abstract = "The emergence of the cloud computing paradigm has greatly enabled innovative service models, such as Platform as a Service (PaaS), and distributed computing frameworks, such as MapReduce. However, most existing cloud systems fail to distinguish users with different preferences, or jobs of different natures. Consequently, they are unable to provide service differentiation, leading to inefficient allocations of cloud resources. Moreover, contentions on the resources exacerbate this inefficiency, when prioritizing crucial jobs is necessary, but impossible. Motivated by this, we propose Abacus, a generic resource management framework addressing this problem. Abacus interacts with users through an auction mechanism, which allows users to specify their priorities using budgets, and job characteristics via utility functions. Based on this information, Abacus computes the optimal allocation and scheduling of resources. Meanwhile, the auction mechanism in Abacus possesses important properties including incentive compatibility (i.e., the users' best strategy is to simply bid their true budgets and job utilities) and monotonicity (i.e., users are motivated to increase their budgets in order to receive better services). In addition, when the user is unclear about her utility function, Abacus automatically learns this function based on statistics of her previous jobs. Extensive experiments, running Hadoop on a private cluster and Amazon EC2, demonstrate the high performance and other desirable properties of Abacus.",
keywords = "Auction, Cloud computing, MapReduce, Service differentiation",
author = "Jianbing Ding and Zhenjie Zhang and Richard, {T. B Ma} and Yin Yang",
year = "2016",
month = "1",
day = "15",
doi = "10.1016/j.comnet.2015.11.007",
language = "English",
volume = "94",
pages = "231--249",
journal = "Computer Networks",
issn = "1389-1286",
publisher = "Elsevier",

}

TY - JOUR

T1 - Auction-based cloud service differentiation with service level objectives

AU - Ding, Jianbing

AU - Zhang, Zhenjie

AU - Richard, T. B Ma

AU - Yang, Yin

PY - 2016/1/15

Y1 - 2016/1/15

N2 - The emergence of the cloud computing paradigm has greatly enabled innovative service models, such as Platform as a Service (PaaS), and distributed computing frameworks, such as MapReduce. However, most existing cloud systems fail to distinguish users with different preferences, or jobs of different natures. Consequently, they are unable to provide service differentiation, leading to inefficient allocations of cloud resources. Moreover, contentions on the resources exacerbate this inefficiency, when prioritizing crucial jobs is necessary, but impossible. Motivated by this, we propose Abacus, a generic resource management framework addressing this problem. Abacus interacts with users through an auction mechanism, which allows users to specify their priorities using budgets, and job characteristics via utility functions. Based on this information, Abacus computes the optimal allocation and scheduling of resources. Meanwhile, the auction mechanism in Abacus possesses important properties including incentive compatibility (i.e., the users' best strategy is to simply bid their true budgets and job utilities) and monotonicity (i.e., users are motivated to increase their budgets in order to receive better services). In addition, when the user is unclear about her utility function, Abacus automatically learns this function based on statistics of her previous jobs. Extensive experiments, running Hadoop on a private cluster and Amazon EC2, demonstrate the high performance and other desirable properties of Abacus.

AB - The emergence of the cloud computing paradigm has greatly enabled innovative service models, such as Platform as a Service (PaaS), and distributed computing frameworks, such as MapReduce. However, most existing cloud systems fail to distinguish users with different preferences, or jobs of different natures. Consequently, they are unable to provide service differentiation, leading to inefficient allocations of cloud resources. Moreover, contentions on the resources exacerbate this inefficiency, when prioritizing crucial jobs is necessary, but impossible. Motivated by this, we propose Abacus, a generic resource management framework addressing this problem. Abacus interacts with users through an auction mechanism, which allows users to specify their priorities using budgets, and job characteristics via utility functions. Based on this information, Abacus computes the optimal allocation and scheduling of resources. Meanwhile, the auction mechanism in Abacus possesses important properties including incentive compatibility (i.e., the users' best strategy is to simply bid their true budgets and job utilities) and monotonicity (i.e., users are motivated to increase their budgets in order to receive better services). In addition, when the user is unclear about her utility function, Abacus automatically learns this function based on statistics of her previous jobs. Extensive experiments, running Hadoop on a private cluster and Amazon EC2, demonstrate the high performance and other desirable properties of Abacus.

KW - Auction

KW - Cloud computing

KW - MapReduce

KW - Service differentiation

UR - http://www.scopus.com/inward/record.url?scp=84954395433&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84954395433&partnerID=8YFLogxK

U2 - 10.1016/j.comnet.2015.11.007

DO - 10.1016/j.comnet.2015.11.007

M3 - Article

AN - SCOPUS:84954395433

VL - 94

SP - 231

EP - 249

JO - Computer Networks

JF - Computer Networks

SN - 1389-1286

ER -