Energy efficiency in office computing environments [Elektronische Ressource] / Andreas Berl

Sujets

• Informatik

Informations

• Publié par: universitat_passau
• Publié le: 01 janvier 2011
• Nombre de lectures: 20
• Langue: English
• Poids de l'ouvrage: 4 Mo

Extrait

Fakulät für Informatik und Mathematik
Universität Passau, Germany
Energy Efficiency in
Office Computing Environments
Andreas Berl
Supervisor: Hermann de Meer
A thesis submitted for
Doctoral Degree
March 20111. Reviewer: Prof. Hermann de Meer
Professor of Computer Networks and Communications
University of Passau
Innstr. 43
94032 Passau, Germany
Email: [e-mail]
Web: http://www.net.fim.uni-passau.de
2. Reviewer: Prof. David Hutchison
Director of InfoLab21 and Professor of Computing
Lancaster University
LA1 4WA
Lancaster, UK
Email: [e-mail]
Web: http://www.infolab21.lancs.ac.ukAbstract
The increasing cost of energy and the worldwide desire to reduce CO emissions2
has raised concern about the energy efficiency of information and communica-
tion technology. Whilst research has focused on data centres recently, this thesis
identifies office computing environments as significant consumers of energy.
Office computing environments offer great potential for energy savings: On one
hand, such environments consist of a large number of hosts. On the other hand,
these hosts often remain turned on 24 hours per day while being underutilised or
even idle. This thesis analyzes the energy consumption within office computing
environments and suggests an energy-efficient virtualized office environment. The
office environment is virtualized to achieve flexible virtualized office resources
that enable an energy-based resource management. This resource management
stops idle services and idle hosts from consuming resources within the office and
consolidates utilised office services on office hosts. This increases the utilisation
of some hosts while other hosts are turned off to save energy. The suggested
architecture is based on a decentralized approach that can be applied to all kinds
of office computing environments, even if no centralized data centre infrastructure
is available.
The thesis develops the architecture of the virtualized office environment together
with an energy consumption model that is able to estimate the energy consumption
of hosts and network within office environments. The model enables the energy-
related comparison of ordinary and virtualized office environments, considering
the energy-efficient management of services.
Furthermore, this thesis evaluates energy efficiency and overhead of the suggested
approach. First, it theoretically proves the energy efficiency of the virtualized of-
fice environment with respect to the energy consumption model. Second, it uses
iiiMarkov processes to evaluate the impact of user behaviour on the suggested ar-
chitecture. Finally, the thesis develops a discrete-event simulation that enables the
simulation and evaluation of office computing environments with respect to vary-
ing virtualization approaches, resource management parameters, user behaviour,
and office equipment. The evaluation shows that the virtualized office environ-
ment saves more than half of the energy consumption within office computing
environments, depending on user behaviour and office equipment.
ivAcknowledgements
I would like to express my sincere gratitude to my supervisor, Prof. Hermann
de Meer, for the continuous support, guidance, and valuable feedback I received
during the last few years. He introduced me to the world of research, entrusted me
with many interesting and challenging tasks, and taught me how to successfully
write scientific publications.
Furthermore, I would like to thank Prof. David Hutchison for enabling my four
months DAAD scholarship at Lancaster University in 2009. He heavily supported
my research with valuable discussions and brought me in contact with researchers
of related fields. Especially, I want to thank Dr. Andreas Mauthe for his excellent
supervision during my time in Lancaster and Gareth Tyson for most interesting dis-
cussions and giving me guided tours. I also want to thank Dr. Nicholas
Race and Dr. Johnathan Ishmael for their help and cooperation.
My special thanks go to all of my colleagues at the University of Passau. Specif-
ically, I want to thank Andreas Fischer and Patrick Wüchner for their support,
insightful discussions, and reviews of my thesis.
Finally, I want to thank my wife Saskia for her support during the last two years,
endless discussions, and helpful reviews.
vContents
List of Figures xi
List of Tables xiii
List of Definitions xv
List of Acronyms xvii
List of Symbols xix
1 Introduction 1
1.1 Energy consumption within office computing environments . . . . . . . . . . . 1
1.2 Solution approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Contribution of this thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Thesis structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2 Related Work 7
2.1 Virtualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.1 Host virtualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.2 Network . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2 Energy efficiency in distributed environments . . . . . . . . . . . . . . . . . . 16
2.2.1 Power-management features of hosts . . . . . . . . . . . . . . . . . . 16
2.2.2 Grid and cluster computing . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2.3 Future home environments . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2.4 Internet Suspend/Resume . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2.5 Office power management . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2.6 Terminal servers and virtual desktop infrastructures . . . . . . . . . . . 21
viiCONTENTS
2.2.7 Cloud computing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3 Power consumption models . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3.1 Host models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3.2 Networked architecture models . . . . . . . . . . . . . . . . . . . . . 27
3 Energy-Efficient Office Design 31
3.1 Energy-saving potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.2 Design principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.3 Requirements and challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4 Virtualized Office Environment Architecture 43
4.1 Virtualization of office resources . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.1.1 PDE execution environments . . . . . . . . . . . . . . . . . . . . . . . 44
4.1.2 PDE migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.1.3 Distributed resource management . . . . . . . . . . . . . . . . . . . . 49
4.2 PDE management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.2.1 Energy states of hosts and PDEs . . . . . . . . . . . . . . . . . . . . . 56
4.2.2 Energy-optimal PDE management . . . . . . . . . . . . . . . . . . . . 57
4.2.3 Service-optimal PDE . . . . . . . . . . . . . . . . . . . . 66
4.2.4 Energy and service-aware PDE management . . . . . . . . . . . . . . 70
4.2.5 Energy efficiency and availability trade-off . . . . . . . . . . . . . . . 76
4.3 Resilience and security issues . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
4.4 Architecture overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
5 Energy Consumption Model 83
5.1 Host power consumption model . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.1.1 Power characteristic . . . . . . . . . . . . . . . . . . . . 84
5.1.2 Ordinary host model . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.1.3 Virtualized host model . . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.2 Network power consumption model . . . . . . . . . . . . . . . . . . . . . . . 90
5.2.1 Power characteristic . . . . . . . . . . . . . . . . . . . . 90
5.2.2 Ordinary network model . . . . . . . . . . . . . . . . . . . . . . . . . 91
5.2.3 Virtualized network model . . . . . . . . . . . . . . . . . . . . . . . . 91
5.3 Office energy consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
viiiCONTENTS
5.4 Energy efficiency proof . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
6 Evaluation 99
6.1 User model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
6.2 Markov process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
6.2.1 Work time scenario and modelling . . . . . . . . . . . . . . . . . . . . 103
6.2.2 PDE management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
6.2.3 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
6.2.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
6.3 Discrete-event simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
6.3.1 DES implementation and validation . . . . . . . . . . . . . . . . . . . 111
6.3.2 Parameter settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
6.3.3 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
6.3.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
6.4 Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
6.4.1 Power meter and systems under te