Is the Same Instance Type Created Equal? Exploiting Heterogeneity of Public Clouds

Zhonghong Ou, Hao Zhuang, Andrey Lukyanenko, Jukka K. Nurminen, Pan Hui, Vladimir Mazalov, Antti Yla-Jaaski

Research output: Contribution to journalArticleScientificpeer-review

37 Citations (Scopus)

Abstract

Public cloud platforms might start with homogeneous hardware; nevertheless, because of inevitable hardware upgrades, or adding more capacity, the initial homogeneous platform will gradually evolve into heterogeneous as time passes by. The consequent performance heterogeneity is of concern to cloud users. In this paper, we evaluate performance variations from hardware heterogeneity and scheduling mechanisms of public clouds. Amazon Elastic Compute Cloud (Amazon EC2) and Rackspace Cloud are used as the representatives because of their relatively long record and wide usage among small and medium enterprises (SMEs). A comprehensive set of microbenchmarks and application-level macrobenchmarks have been used to investigate performance variation. Several major contributions have been made. First, we find out that heterogeneous hardware is a commonality among the relatively long-lasting cloud platforms, although the level of heterogeneity varies. Second, we observe that heterogeneous hardware is the primary culprit of performance variation of cloud platforms. Third, we discover that varied CPU acquisition percentages and different virtual machine scheduling mechanisms exacerbate the performance variation problem, especially for network related operations. Finally, based on the observations, we propose cost-saving approaches and analyze Nash equilibrium from cloud user perspective. By using a simple "trial-and-better" approach, i.e., keep good-performing instances and discard bad-performing instances, cloud users can achieve up to 30 percent cost saving.
Original languageEnglish
Pages (from-to)201-214
Number of pages14
JournalIEEE Transactions on Cloud Computing
Volume1
Issue number2
DOIs
Publication statusPublished - 1 Jul 2013
MoE publication typeA1 Journal article-refereed

Fingerprint

Hardware
Scheduling
Program processors
Costs
Industry

Keywords

  • Amazon EC2
  • Hardware heterogeneity
  • VM scheduling mechanism
  • cloud computing
  • performance variation

Cite this

Ou, Z., Zhuang, H., Lukyanenko, A., Nurminen, J. K., Hui, P., Mazalov, V., & Yla-Jaaski, A. (2013). Is the Same Instance Type Created Equal? Exploiting Heterogeneity of Public Clouds. IEEE Transactions on Cloud Computing, 1(2), 201-214. https://doi.org/10.1109/TCC.2013.12
Ou, Zhonghong ; Zhuang, Hao ; Lukyanenko, Andrey ; Nurminen, Jukka K. ; Hui, Pan ; Mazalov, Vladimir ; Yla-Jaaski, Antti. / Is the Same Instance Type Created Equal? Exploiting Heterogeneity of Public Clouds. In: IEEE Transactions on Cloud Computing. 2013 ; Vol. 1, No. 2. pp. 201-214.
@article{b2c03db5065b4a5a8ddb20261358947b,
title = "Is the Same Instance Type Created Equal? Exploiting Heterogeneity of Public Clouds",
abstract = "Public cloud platforms might start with homogeneous hardware; nevertheless, because of inevitable hardware upgrades, or adding more capacity, the initial homogeneous platform will gradually evolve into heterogeneous as time passes by. The consequent performance heterogeneity is of concern to cloud users. In this paper, we evaluate performance variations from hardware heterogeneity and scheduling mechanisms of public clouds. Amazon Elastic Compute Cloud (Amazon EC2) and Rackspace Cloud are used as the representatives because of their relatively long record and wide usage among small and medium enterprises (SMEs). A comprehensive set of microbenchmarks and application-level macrobenchmarks have been used to investigate performance variation. Several major contributions have been made. First, we find out that heterogeneous hardware is a commonality among the relatively long-lasting cloud platforms, although the level of heterogeneity varies. Second, we observe that heterogeneous hardware is the primary culprit of performance variation of cloud platforms. Third, we discover that varied CPU acquisition percentages and different virtual machine scheduling mechanisms exacerbate the performance variation problem, especially for network related operations. Finally, based on the observations, we propose cost-saving approaches and analyze Nash equilibrium from cloud user perspective. By using a simple {"}trial-and-better{"} approach, i.e., keep good-performing instances and discard bad-performing instances, cloud users can achieve up to 30 percent cost saving.",
keywords = "Amazon EC2, Hardware heterogeneity, VM scheduling mechanism, cloud computing, performance variation",
author = "Zhonghong Ou and Hao Zhuang and Andrey Lukyanenko and Nurminen, {Jukka K.} and Pan Hui and Vladimir Mazalov and Antti Yla-Jaaski",
year = "2013",
month = "7",
day = "1",
doi = "10.1109/TCC.2013.12",
language = "English",
volume = "1",
pages = "201--214",
journal = "IEEE Transactions on Cloud Computing",
issn = "2168-7161",
publisher = "Institute of Electrical and Electronic Engineers IEEE",
number = "2",

}

Ou, Z, Zhuang, H, Lukyanenko, A, Nurminen, JK, Hui, P, Mazalov, V & Yla-Jaaski, A 2013, 'Is the Same Instance Type Created Equal? Exploiting Heterogeneity of Public Clouds', IEEE Transactions on Cloud Computing, vol. 1, no. 2, pp. 201-214. https://doi.org/10.1109/TCC.2013.12

Is the Same Instance Type Created Equal? Exploiting Heterogeneity of Public Clouds. / Ou, Zhonghong; Zhuang, Hao; Lukyanenko, Andrey; Nurminen, Jukka K.; Hui, Pan; Mazalov, Vladimir; Yla-Jaaski, Antti.

In: IEEE Transactions on Cloud Computing, Vol. 1, No. 2, 01.07.2013, p. 201-214.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Is the Same Instance Type Created Equal? Exploiting Heterogeneity of Public Clouds

AU - Ou, Zhonghong

AU - Zhuang, Hao

AU - Lukyanenko, Andrey

AU - Nurminen, Jukka K.

AU - Hui, Pan

AU - Mazalov, Vladimir

AU - Yla-Jaaski, Antti

PY - 2013/7/1

Y1 - 2013/7/1

N2 - Public cloud platforms might start with homogeneous hardware; nevertheless, because of inevitable hardware upgrades, or adding more capacity, the initial homogeneous platform will gradually evolve into heterogeneous as time passes by. The consequent performance heterogeneity is of concern to cloud users. In this paper, we evaluate performance variations from hardware heterogeneity and scheduling mechanisms of public clouds. Amazon Elastic Compute Cloud (Amazon EC2) and Rackspace Cloud are used as the representatives because of their relatively long record and wide usage among small and medium enterprises (SMEs). A comprehensive set of microbenchmarks and application-level macrobenchmarks have been used to investigate performance variation. Several major contributions have been made. First, we find out that heterogeneous hardware is a commonality among the relatively long-lasting cloud platforms, although the level of heterogeneity varies. Second, we observe that heterogeneous hardware is the primary culprit of performance variation of cloud platforms. Third, we discover that varied CPU acquisition percentages and different virtual machine scheduling mechanisms exacerbate the performance variation problem, especially for network related operations. Finally, based on the observations, we propose cost-saving approaches and analyze Nash equilibrium from cloud user perspective. By using a simple "trial-and-better" approach, i.e., keep good-performing instances and discard bad-performing instances, cloud users can achieve up to 30 percent cost saving.

AB - Public cloud platforms might start with homogeneous hardware; nevertheless, because of inevitable hardware upgrades, or adding more capacity, the initial homogeneous platform will gradually evolve into heterogeneous as time passes by. The consequent performance heterogeneity is of concern to cloud users. In this paper, we evaluate performance variations from hardware heterogeneity and scheduling mechanisms of public clouds. Amazon Elastic Compute Cloud (Amazon EC2) and Rackspace Cloud are used as the representatives because of their relatively long record and wide usage among small and medium enterprises (SMEs). A comprehensive set of microbenchmarks and application-level macrobenchmarks have been used to investigate performance variation. Several major contributions have been made. First, we find out that heterogeneous hardware is a commonality among the relatively long-lasting cloud platforms, although the level of heterogeneity varies. Second, we observe that heterogeneous hardware is the primary culprit of performance variation of cloud platforms. Third, we discover that varied CPU acquisition percentages and different virtual machine scheduling mechanisms exacerbate the performance variation problem, especially for network related operations. Finally, based on the observations, we propose cost-saving approaches and analyze Nash equilibrium from cloud user perspective. By using a simple "trial-and-better" approach, i.e., keep good-performing instances and discard bad-performing instances, cloud users can achieve up to 30 percent cost saving.

KW - Amazon EC2

KW - Hardware heterogeneity

KW - VM scheduling mechanism

KW - cloud computing

KW - performance variation

U2 - 10.1109/TCC.2013.12

DO - 10.1109/TCC.2013.12

M3 - Article

VL - 1

SP - 201

EP - 214

JO - IEEE Transactions on Cloud Computing

JF - IEEE Transactions on Cloud Computing

SN - 2168-7161

IS - 2

ER -