A dynamic energy management scheme for multi-tier data centers

Seung Hwan Lim, Bikash Sharma, Byung Chul Tak, Chitaranjan Das

Research output: Chapter in Book/Report/Conference proceedingConference contribution

16 Citations (Scopus)

Abstract

Multi-tier data centers have become a norm for hosting modern Internet applications because they provide a flexible, modular, scalable and high performance environment. However, these benefits come at a price of the economic dent incurred in powering and cooling these large hosting centers. Thus, energy efficiency has become a critical consideration in designing Internet data centers. In this paper, we propose a multifaceted approach, Hybrid, consisting of dynamic provisioning, frequency scaling and dynamic power management (DPM) schemes to reduce the energy consumption of multi-tier data centers, while meeting the Service Level Agreements (SLAs). We formulate a mathematical model of the energy and performance/SLA optimization problem followed by a queueing theory based approach to develop two heuristics for solving the optimization problem. The first heuristic dynamically provisions the optimal number of servers required in each tier. The second heuristic proactively decides the CPU speed and the duration of sleep states of a server to achieve further energy savings. We evaluate our heuristics using a simulator that was validated with real measurements on a prototype three-tier data center consisting of 25 servers with two multi-tier application benchmarks. Our experimental results indicate that the proposed scheme, Hybrid, can reduce the energy consumption by 50% relative to static provisioning without CPU frequency scaling and DPM. We demonstrate that Hybrid satisfies the SLAs for dynamically varying workloads. In addition, the proposed multifaceted approach is more energy efficient than the other methods such as dynamic provisioning with exploiting deep sleep states.

Original languageEnglish (US)
Title of host publicationISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software
Pages257-266
Number of pages10
DOIs
StatePublished - May 30 2011
EventIEEE International Symposium on Performance Analysis of Systems and Software, ISPASS 2011 - Austin, TX, United States
Duration: Apr 10 2011Apr 12 2011

Publication series

NameISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software

Other

OtherIEEE International Symposium on Performance Analysis of Systems and Software, ISPASS 2011
CountryUnited States
CityAustin, TX
Period4/10/114/12/11

Fingerprint

Energy management
Servers
Program processors
Energy utilization
Internet
Queueing theory
Energy efficiency
Energy conservation
Simulators
Mathematical models
Cooling
Economics
Power management
Sleep

All Science Journal Classification (ASJC) codes

  • Software

Cite this

Lim, S. H., Sharma, B., Tak, B. C., & Das, C. (2011). A dynamic energy management scheme for multi-tier data centers. In ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software (pp. 257-266). [5762742] (ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software). https://doi.org/10.1109/ISPASS.2011.5762742
Lim, Seung Hwan ; Sharma, Bikash ; Tak, Byung Chul ; Das, Chitaranjan. / A dynamic energy management scheme for multi-tier data centers. ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software. 2011. pp. 257-266 (ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software).
@inproceedings{191fdd7e2fd14e1c8e1e57ae7e720d02,
title = "A dynamic energy management scheme for multi-tier data centers",
abstract = "Multi-tier data centers have become a norm for hosting modern Internet applications because they provide a flexible, modular, scalable and high performance environment. However, these benefits come at a price of the economic dent incurred in powering and cooling these large hosting centers. Thus, energy efficiency has become a critical consideration in designing Internet data centers. In this paper, we propose a multifaceted approach, Hybrid, consisting of dynamic provisioning, frequency scaling and dynamic power management (DPM) schemes to reduce the energy consumption of multi-tier data centers, while meeting the Service Level Agreements (SLAs). We formulate a mathematical model of the energy and performance/SLA optimization problem followed by a queueing theory based approach to develop two heuristics for solving the optimization problem. The first heuristic dynamically provisions the optimal number of servers required in each tier. The second heuristic proactively decides the CPU speed and the duration of sleep states of a server to achieve further energy savings. We evaluate our heuristics using a simulator that was validated with real measurements on a prototype three-tier data center consisting of 25 servers with two multi-tier application benchmarks. Our experimental results indicate that the proposed scheme, Hybrid, can reduce the energy consumption by 50{\%} relative to static provisioning without CPU frequency scaling and DPM. We demonstrate that Hybrid satisfies the SLAs for dynamically varying workloads. In addition, the proposed multifaceted approach is more energy efficient than the other methods such as dynamic provisioning with exploiting deep sleep states.",
author = "Lim, {Seung Hwan} and Bikash Sharma and Tak, {Byung Chul} and Chitaranjan Das",
year = "2011",
month = "5",
day = "30",
doi = "10.1109/ISPASS.2011.5762742",
language = "English (US)",
isbn = "9781612843681",
series = "ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software",
pages = "257--266",
booktitle = "ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software",

}

Lim, SH, Sharma, B, Tak, BC & Das, C 2011, A dynamic energy management scheme for multi-tier data centers. in ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software., 5762742, ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software, pp. 257-266, IEEE International Symposium on Performance Analysis of Systems and Software, ISPASS 2011, Austin, TX, United States, 4/10/11. https://doi.org/10.1109/ISPASS.2011.5762742

A dynamic energy management scheme for multi-tier data centers. / Lim, Seung Hwan; Sharma, Bikash; Tak, Byung Chul; Das, Chitaranjan.

ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software. 2011. p. 257-266 5762742 (ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - A dynamic energy management scheme for multi-tier data centers

AU - Lim, Seung Hwan

AU - Sharma, Bikash

AU - Tak, Byung Chul

AU - Das, Chitaranjan

PY - 2011/5/30

Y1 - 2011/5/30

N2 - Multi-tier data centers have become a norm for hosting modern Internet applications because they provide a flexible, modular, scalable and high performance environment. However, these benefits come at a price of the economic dent incurred in powering and cooling these large hosting centers. Thus, energy efficiency has become a critical consideration in designing Internet data centers. In this paper, we propose a multifaceted approach, Hybrid, consisting of dynamic provisioning, frequency scaling and dynamic power management (DPM) schemes to reduce the energy consumption of multi-tier data centers, while meeting the Service Level Agreements (SLAs). We formulate a mathematical model of the energy and performance/SLA optimization problem followed by a queueing theory based approach to develop two heuristics for solving the optimization problem. The first heuristic dynamically provisions the optimal number of servers required in each tier. The second heuristic proactively decides the CPU speed and the duration of sleep states of a server to achieve further energy savings. We evaluate our heuristics using a simulator that was validated with real measurements on a prototype three-tier data center consisting of 25 servers with two multi-tier application benchmarks. Our experimental results indicate that the proposed scheme, Hybrid, can reduce the energy consumption by 50% relative to static provisioning without CPU frequency scaling and DPM. We demonstrate that Hybrid satisfies the SLAs for dynamically varying workloads. In addition, the proposed multifaceted approach is more energy efficient than the other methods such as dynamic provisioning with exploiting deep sleep states.

AB - Multi-tier data centers have become a norm for hosting modern Internet applications because they provide a flexible, modular, scalable and high performance environment. However, these benefits come at a price of the economic dent incurred in powering and cooling these large hosting centers. Thus, energy efficiency has become a critical consideration in designing Internet data centers. In this paper, we propose a multifaceted approach, Hybrid, consisting of dynamic provisioning, frequency scaling and dynamic power management (DPM) schemes to reduce the energy consumption of multi-tier data centers, while meeting the Service Level Agreements (SLAs). We formulate a mathematical model of the energy and performance/SLA optimization problem followed by a queueing theory based approach to develop two heuristics for solving the optimization problem. The first heuristic dynamically provisions the optimal number of servers required in each tier. The second heuristic proactively decides the CPU speed and the duration of sleep states of a server to achieve further energy savings. We evaluate our heuristics using a simulator that was validated with real measurements on a prototype three-tier data center consisting of 25 servers with two multi-tier application benchmarks. Our experimental results indicate that the proposed scheme, Hybrid, can reduce the energy consumption by 50% relative to static provisioning without CPU frequency scaling and DPM. We demonstrate that Hybrid satisfies the SLAs for dynamically varying workloads. In addition, the proposed multifaceted approach is more energy efficient than the other methods such as dynamic provisioning with exploiting deep sleep states.

UR - http://www.scopus.com/inward/record.url?scp=79957504751&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79957504751&partnerID=8YFLogxK

U2 - 10.1109/ISPASS.2011.5762742

DO - 10.1109/ISPASS.2011.5762742

M3 - Conference contribution

AN - SCOPUS:79957504751

SN - 9781612843681

T3 - ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software

SP - 257

EP - 266

BT - ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software

ER -

Lim SH, Sharma B, Tak BC, Das C. A dynamic energy management scheme for multi-tier data centers. In ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software. 2011. p. 257-266. 5762742. (ISPASS 2011 - IEEE International Symposium on Performance Analysis of Systems and Software). https://doi.org/10.1109/ISPASS.2011.5762742