Software-directed data access scheduling for reducing disk energy consumption

Yuanrui Zhang, Jun Liu, Mahmut Kandemir

Research output: Contribution to conferencePaper

4 Citations (Scopus)

Abstract

Most existing research in disk power management has focused on exploiting idle periods of disks. Both hardware power-saving mechanisms (such as spin-down disks and multi-speed disks) and complementary software strategies (such as code and data layout transformations to increase the length of idle periods) have been explored. However, while hardware power-saving mechanisms cannot handle short idle periods of high-performance parallel applications, prior code/data reorganization strategies typically require extensive code modifications. In this paper, we propose and evaluate a compiler-directed data access (I/O call) scheduling framework for saving disk energy, which groups as many data requests as possible in a shorter period, thus creating longer disk idle periods for improving the effectiveness of hardware power-saving mechanisms. As compared to prior software based efforts, it requires no code or data restructuring. We evaluate our approach using six application programs in a cluster-based simulation environment. The experimental results show that it improves the effectiveness of both spin-down disks and multispeed disks with doubled power savings on average.

Original languageEnglish (US)
Pages596-605
Number of pages10
DOIs
StatePublished - Oct 5 2012
Event32nd IEEE International Conference on Distributed Computing Systems, ICDCS 2012 - Macau, China
Duration: Jun 18 2012Jun 21 2012

Other

Other32nd IEEE International Conference on Distributed Computing Systems, ICDCS 2012
CountryChina
CityMacau
Period6/18/126/21/12

Fingerprint

Energy utilization
Scheduling
Hardware
Application programs
Power management

All Science Journal Classification (ASJC) codes

  • Software
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

Zhang, Y., Liu, J., & Kandemir, M. (2012). Software-directed data access scheduling for reducing disk energy consumption. 596-605. Paper presented at 32nd IEEE International Conference on Distributed Computing Systems, ICDCS 2012, Macau, China. https://doi.org/10.1109/ICDCS.2012.12
Zhang, Yuanrui ; Liu, Jun ; Kandemir, Mahmut. / Software-directed data access scheduling for reducing disk energy consumption. Paper presented at 32nd IEEE International Conference on Distributed Computing Systems, ICDCS 2012, Macau, China.10 p.
@conference{b6ce4621dabf45a28928f4ffb8973201,
title = "Software-directed data access scheduling for reducing disk energy consumption",
abstract = "Most existing research in disk power management has focused on exploiting idle periods of disks. Both hardware power-saving mechanisms (such as spin-down disks and multi-speed disks) and complementary software strategies (such as code and data layout transformations to increase the length of idle periods) have been explored. However, while hardware power-saving mechanisms cannot handle short idle periods of high-performance parallel applications, prior code/data reorganization strategies typically require extensive code modifications. In this paper, we propose and evaluate a compiler-directed data access (I/O call) scheduling framework for saving disk energy, which groups as many data requests as possible in a shorter period, thus creating longer disk idle periods for improving the effectiveness of hardware power-saving mechanisms. As compared to prior software based efforts, it requires no code or data restructuring. We evaluate our approach using six application programs in a cluster-based simulation environment. The experimental results show that it improves the effectiveness of both spin-down disks and multispeed disks with doubled power savings on average.",
author = "Yuanrui Zhang and Jun Liu and Mahmut Kandemir",
year = "2012",
month = "10",
day = "5",
doi = "10.1109/ICDCS.2012.12",
language = "English (US)",
pages = "596--605",
note = "32nd IEEE International Conference on Distributed Computing Systems, ICDCS 2012 ; Conference date: 18-06-2012 Through 21-06-2012",

}

Zhang, Y, Liu, J & Kandemir, M 2012, 'Software-directed data access scheduling for reducing disk energy consumption' Paper presented at 32nd IEEE International Conference on Distributed Computing Systems, ICDCS 2012, Macau, China, 6/18/12 - 6/21/12, pp. 596-605. https://doi.org/10.1109/ICDCS.2012.12

Software-directed data access scheduling for reducing disk energy consumption. / Zhang, Yuanrui; Liu, Jun; Kandemir, Mahmut.

2012. 596-605 Paper presented at 32nd IEEE International Conference on Distributed Computing Systems, ICDCS 2012, Macau, China.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Software-directed data access scheduling for reducing disk energy consumption

AU - Zhang, Yuanrui

AU - Liu, Jun

AU - Kandemir, Mahmut

PY - 2012/10/5

Y1 - 2012/10/5

N2 - Most existing research in disk power management has focused on exploiting idle periods of disks. Both hardware power-saving mechanisms (such as spin-down disks and multi-speed disks) and complementary software strategies (such as code and data layout transformations to increase the length of idle periods) have been explored. However, while hardware power-saving mechanisms cannot handle short idle periods of high-performance parallel applications, prior code/data reorganization strategies typically require extensive code modifications. In this paper, we propose and evaluate a compiler-directed data access (I/O call) scheduling framework for saving disk energy, which groups as many data requests as possible in a shorter period, thus creating longer disk idle periods for improving the effectiveness of hardware power-saving mechanisms. As compared to prior software based efforts, it requires no code or data restructuring. We evaluate our approach using six application programs in a cluster-based simulation environment. The experimental results show that it improves the effectiveness of both spin-down disks and multispeed disks with doubled power savings on average.

AB - Most existing research in disk power management has focused on exploiting idle periods of disks. Both hardware power-saving mechanisms (such as spin-down disks and multi-speed disks) and complementary software strategies (such as code and data layout transformations to increase the length of idle periods) have been explored. However, while hardware power-saving mechanisms cannot handle short idle periods of high-performance parallel applications, prior code/data reorganization strategies typically require extensive code modifications. In this paper, we propose and evaluate a compiler-directed data access (I/O call) scheduling framework for saving disk energy, which groups as many data requests as possible in a shorter period, thus creating longer disk idle periods for improving the effectiveness of hardware power-saving mechanisms. As compared to prior software based efforts, it requires no code or data restructuring. We evaluate our approach using six application programs in a cluster-based simulation environment. The experimental results show that it improves the effectiveness of both spin-down disks and multispeed disks with doubled power savings on average.

UR - http://www.scopus.com/inward/record.url?scp=84866954001&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866954001&partnerID=8YFLogxK

U2 - 10.1109/ICDCS.2012.12

DO - 10.1109/ICDCS.2012.12

M3 - Paper

AN - SCOPUS:84866954001

SP - 596

EP - 605

ER -

Zhang Y, Liu J, Kandemir M. Software-directed data access scheduling for reducing disk energy consumption. 2012. Paper presented at 32nd IEEE International Conference on Distributed Computing Systems, ICDCS 2012, Macau, China. https://doi.org/10.1109/ICDCS.2012.12