Energy-efficient storage in virtual machine environments

Lei Ye, Gen Lu, Sushanth Kumar, Christopher Gniady, John H Hartman

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Current trends in increasing storage capacity and virtualization of resources combined with the need for energy efficiency put a challenging task in front of system designers. Previous studies have suggested many approaches to reduce hard disk energy dissipation in native OS environments; however, those mechanisms do not perform well in virtual machine environments because a virtual machine (VM) and the virtual machine monitor (VMM) that runs it have different semantic contexts. This paper explores the disk I/O activities between VMM and VMs using trace driven simulation to understand the I/O behavior of the VM system. Subsequently, this paper proposes three mechanisms to address the isolation between VMM and VMs, and increase the burstiness of hard disk accesses to increase energy efficiency of a hard disk. Compared to standard shutdown mechanisms, with eight VMs the proposed mechanisms reduce disk spin-ups, increase the disk sleep time, and reduce energy consumption by 14.8% with only 0.5% increase in execution time. We implemented the proposed mechanisms in Xen and validated our simulation results.

Original languageEnglish (US)
Pages (from-to)75-84
Number of pages10
JournalACM SIGPLAN Notices
Volume45
Issue number7
DOIs
StatePublished - Jul 2010

Fingerprint

Hard disk storage
Energy efficiency
Computer monitors
Virtual machine
Energy dissipation
Energy utilization
Semantics

Keywords

  • Energy Management
  • Storage System
  • Virtual Machine

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Energy-efficient storage in virtual machine environments. / Ye, Lei; Lu, Gen; Kumar, Sushanth; Gniady, Christopher; Hartman, John H.

In: ACM SIGPLAN Notices, Vol. 45, No. 7, 07.2010, p. 75-84.

Research output: Contribution to journalArticle

Ye, Lei ; Lu, Gen ; Kumar, Sushanth ; Gniady, Christopher ; Hartman, John H. / Energy-efficient storage in virtual machine environments. In: ACM SIGPLAN Notices. 2010 ; Vol. 45, No. 7. pp. 75-84.
@article{f2b96f67ff604418aa37e24ee2a5582f,
title = "Energy-efficient storage in virtual machine environments",
abstract = "Current trends in increasing storage capacity and virtualization of resources combined with the need for energy efficiency put a challenging task in front of system designers. Previous studies have suggested many approaches to reduce hard disk energy dissipation in native OS environments; however, those mechanisms do not perform well in virtual machine environments because a virtual machine (VM) and the virtual machine monitor (VMM) that runs it have different semantic contexts. This paper explores the disk I/O activities between VMM and VMs using trace driven simulation to understand the I/O behavior of the VM system. Subsequently, this paper proposes three mechanisms to address the isolation between VMM and VMs, and increase the burstiness of hard disk accesses to increase energy efficiency of a hard disk. Compared to standard shutdown mechanisms, with eight VMs the proposed mechanisms reduce disk spin-ups, increase the disk sleep time, and reduce energy consumption by 14.8{\%} with only 0.5{\%} increase in execution time. We implemented the proposed mechanisms in Xen and validated our simulation results.",
keywords = "Energy Management, Storage System, Virtual Machine",
author = "Lei Ye and Gen Lu and Sushanth Kumar and Christopher Gniady and Hartman, {John H}",
year = "2010",
month = "7",
doi = "10.1145/1837854.1736009",
language = "English (US)",
volume = "45",
pages = "75--84",
journal = "ACM SIGPLAN Notices",
issn = "1523-2867",
publisher = "Association for Computing Machinery (ACM)",
number = "7",

}

TY - JOUR

T1 - Energy-efficient storage in virtual machine environments

AU - Ye, Lei

AU - Lu, Gen

AU - Kumar, Sushanth

AU - Gniady, Christopher

AU - Hartman, John H

PY - 2010/7

Y1 - 2010/7

N2 - Current trends in increasing storage capacity and virtualization of resources combined with the need for energy efficiency put a challenging task in front of system designers. Previous studies have suggested many approaches to reduce hard disk energy dissipation in native OS environments; however, those mechanisms do not perform well in virtual machine environments because a virtual machine (VM) and the virtual machine monitor (VMM) that runs it have different semantic contexts. This paper explores the disk I/O activities between VMM and VMs using trace driven simulation to understand the I/O behavior of the VM system. Subsequently, this paper proposes three mechanisms to address the isolation between VMM and VMs, and increase the burstiness of hard disk accesses to increase energy efficiency of a hard disk. Compared to standard shutdown mechanisms, with eight VMs the proposed mechanisms reduce disk spin-ups, increase the disk sleep time, and reduce energy consumption by 14.8% with only 0.5% increase in execution time. We implemented the proposed mechanisms in Xen and validated our simulation results.

AB - Current trends in increasing storage capacity and virtualization of resources combined with the need for energy efficiency put a challenging task in front of system designers. Previous studies have suggested many approaches to reduce hard disk energy dissipation in native OS environments; however, those mechanisms do not perform well in virtual machine environments because a virtual machine (VM) and the virtual machine monitor (VMM) that runs it have different semantic contexts. This paper explores the disk I/O activities between VMM and VMs using trace driven simulation to understand the I/O behavior of the VM system. Subsequently, this paper proposes three mechanisms to address the isolation between VMM and VMs, and increase the burstiness of hard disk accesses to increase energy efficiency of a hard disk. Compared to standard shutdown mechanisms, with eight VMs the proposed mechanisms reduce disk spin-ups, increase the disk sleep time, and reduce energy consumption by 14.8% with only 0.5% increase in execution time. We implemented the proposed mechanisms in Xen and validated our simulation results.

KW - Energy Management

KW - Storage System

KW - Virtual Machine

UR - http://www.scopus.com/inward/record.url?scp=79959890438&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79959890438&partnerID=8YFLogxK

U2 - 10.1145/1837854.1736009

DO - 10.1145/1837854.1736009

M3 - Article

AN - SCOPUS:79959890438

VL - 45

SP - 75

EP - 84

JO - ACM SIGPLAN Notices

JF - ACM SIGPLAN Notices

SN - 1523-2867

IS - 7

ER -