Muutke küpsiste eelistusi

E-raamat: Energy-Efficient Distributed Computing Systems

Edited by , Edited by
Teised raamatud teemal:
  • Formaat - PDF+DRM
  • Hind: 154,31 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Lisa ostukorvi
  • Lisa soovinimekirja
  • See e-raamat on mõeldud ainult isiklikuks kasutamiseks. E-raamatuid ei saa tagastada.
  • Raamatukogudele
Energy-Efficient Distributed Computing SystemsSuurem pilt
Teised raamatud teemal:

DRM piirangud

  • Kopeerimine (copy/paste):

    ei ole lubatud

  • Printimine:

    ei ole lubatud

  • Kasutamine:

    Digitaalõiguste kaitse (DRM)
    Kirjastus on väljastanud selle e-raamatu krüpteeritud kujul, mis tähendab, et selle lugemiseks peate installeerima spetsiaalse tarkvara. Samuti peate looma endale  Adobe ID Rohkem infot siin. E-raamatut saab lugeda 1 kasutaja ning alla laadida kuni 6'de seadmesse (kõik autoriseeritud sama Adobe ID-ga).

    Vajalik tarkvara
    Mobiilsetes seadmetes (telefon või tahvelarvuti) lugemiseks peate installeerima selle tasuta rakenduse: PocketBook Reader (iOS / Android)

    PC või Mac seadmes lugemiseks peate installima Adobe Digital Editionsi (Seeon tasuta rakendus spetsiaalselt e-raamatute lugemiseks. Seda ei tohi segamini ajada Adober Reader'iga, mis tõenäoliselt on juba teie arvutisse installeeritud )

    Seda e-raamatut ei saa lugeda Amazon Kindle's. 

The energy consumption issue in distributed computing systems raises various monetary, environmental and system performance concerns. Electricity consumption in the US doubled from 2000 to 2005.  From a financial and environmental standpoint, reducing the consumption of electricity is important, yet these reforms must not lead to performance degradation of the computing systems.  These contradicting constraints create a suite of complex problems that need to be resolved in order to lead to greener distributed computing systems.  This book brings together a group of outstanding researchers that investigate the different facets of green and energy efficient distributed computing.Key features:One of the first books of its kind Features latest research findings on emerging topics by well-known scientists Valuable research for grad students, postdocs, and researchers Research will greatly feed into other technologies and application domains
Preface xxix
Acknowledgments xxxi
Contributors xxxiii
1 Power Allocation And Task Scheduling On Multiprocessor Computers With Energy And Time Constraints
1(38)
Keqin Li
1.1 Introduction
1(4)
1.1.1 Energy Consumption
1(1)
1.1.2 Power Reduction
2(1)
1.1.3 Dynamic Power Management
3(1)
1.1.4 Task Scheduling with Energy and Time Constraints
4(1)
1.1.5
Chapter Outline
5(1)
1.2 Preliminaries
5(5)
1.2.1 Power Consumption Model
5(1)
1.2.2 Problem Definitions
6(1)
1.2.3 Task Models
7(1)
1.2.4 Processor Models
8(1)
1.2.5 Scheduling Models
9(1)
1.2.6 Problem Decomposition
9(1)
1.2.7 Types of Algorithms
10(1)
1.3 Problem Analysis
10(6)
1.3.1 Schedule Length Minimization
10(1)
1.3.1.1 Uniprocessor computers
10(1)
1.3.1.2 Multiprocessor computers
11(1)
1.3.2 Energy Consumption Minimization
12(1)
1.3.2.1 Uniprocessor computers
12(1)
1.3.2.2 Multiprocessor computers
13(1)
1.3.3 Strong NP-Hardness
14(1)
1.3.4 Lower Bounds
14(1)
1.3.5 Energy-Delay Trade-off
15(1)
1.4 Pre-Power-Determination Algorithms
16(12)
1.4.1 Overview
16(1)
1.4.2 Performance Measures
17(1)
1.4.3 Equal-Time Algorithms and Analysis
18(1)
1.4.3.1 Schedule length minimization
18(1)
1.4.3.2 Energy consumption minimization
19(1)
1.4.4 Equal-Energy Algorithms and Analysis
19(1)
1.4.4.1 Schedule length minimization
19(2)
1.4.4.2 Energy consumption minimization
21(1)
1.4.5 Equal-Speed Algorithms and Analysis
22(1)
1.4.5.1 Schedule length minimization
22(1)
1.4.5.2 Energy consumption minimization
23(1)
1.4.6 Numerical Data
24(1)
1.4.7 Simulation Results
25(3)
1.5 Post-Power-Determination Algorithms
28(5)
1.5.1 Overview
28(1)
1.5.2 Analysis of List Scheduling Algorithms
29(1)
1.5.2.1 Analysis of algorithm LS
29(1)
1.5.2.2 Analysis of algorithm LRF
30(1)
1.5.3 Application to Schedule Length Minimization
30(1)
1.5.4 Application to Energy Consumption Minimization
31(1)
1.5.5 Numerical Data
32(1)
1.5.6 Simulation Results
32(1)
1.6 Summary and Further Research
33(6)
References
34(5)
2 Power-Aware High Performance Computing
39(42)
Rong Ge
Kirk W. Cameron
2.1 Introduction
39(2)
2.2 Background
41(4)
2.2.1 Current Hardware Technology and Power Consumption
41(1)
2.2.1.1 Processor power
41(1)
2.2.1.2 Memory subsystem power
42(1)
2.2.2 Performance
43(1)
2.2.3 Energy Efficiency
44(1)
2.3 Related Work
45(3)
2.3.1 Power Profiling
45(1)
2.3.1.1 Simulator-based power estimation
45(1)
2.3.1.2 Direct measurements
46(1)
2.3.1.3 Event-based estimation
46(1)
2.3.2 Performance Scalability on Power-Aware Systems
46(1)
2.3.3 Adaptive Power Allocation for Energy-Efficient Computing
47(1)
2.4 PowerPack: Fine-Grain Energy Profiling of HPC Applications
48(11)
2.4.1 Design and Implementation of PowerPack
48(1)
2.4.1.1 Overview
48(2)
2.4.1.2 Fine-grain systematic power measurement
50(1)
2.4.1.3 Automatic power profiling and code synchronization
51(2)
2.4.2 Power Profiles of HPC Applications and Systems
53(1)
2.4.2.1 Power distribution over components
53(1)
2.4.2.2 Power dynamics of applications
54(1)
2.4.2.3 Power bounds on HPC systems
55(2)
2.4.2.4 Power versus dynamic voltage and frequency scaling
57(2)
2.5 Power-Aware Speedup Model
59(10)
2.5.1 Power-Aware Speedup
59(1)
2.5.1.1 Sequential execution time for a single workload T1 (w, f)
60(1)
2.5.1.2 Sequential execution time for an ON-chip/OFF-chip workload
60(1)
2.5.1.3 Parallel execution time on N processors for an ON-/OFF-chip workload with DOP = i
61(1)
2.5.1.4 Power-aware speedup for DOP and ON-/OFF-chip workloads
62(1)
2.5.2 Model Parametrization and Validation
63(1)
2.5.2.1 Coarse-grain parametrization and validation
64(2)
2.5.2.2 Fine-grain parametrization and validation
66(3)
2.6 Model Usages
69(4)
2.6.1 Identification of Optimal System Configurations
70(1)
2.6.2 PAS-Directed Energy-Driven Runtime Frequency Scaling
71(2)
2.7 Conclusion
73(8)
References
75(6)
3 Energy Efficiency In HPC Systems
81(28)
Ivan Rodero
Manish Parashar
3.1 Introduction
81(2)
3.2 Background and Related Work
83(5)
3.2.1 CPU Power Management
83(1)
3.2.1.1 OS-level CPU power management
83(1)
3.2.1.2 Workload-level CPU power management
84(1)
3.2.1.3 Cluster-level CPU power management
84(1)
3.2.2 Component-Based Power Management
85(1)
3.2.2.1 Memory subsystem
85(1)
3.2.2.2 Storage subsystem
86(1)
3.2.3 Thermal-Conscious Power Management
87(1)
3.2.4 Power Management in Virtualized Datacenters
87(1)
3.3 Proactive, Component-Based Power Management
88(3)
3.3.1 Job Allocation Policies
88(2)
3.3.2 Workload Profiling
90(1)
3.4 Quantifying Energy Saving Possibilities
91(4)
3.4.1 Methodology
92(1)
3.4.2 Component-Level Power Requirements
92(2)
3.4.3 Energy Savings
94(1)
3.5 Evaluation of the Proposed Strategies
95(2)
3.5.1 Methodology
96(1)
3.5.2 Workloads
96(1)
3.5.3 Metrics
97(1)
3.6 Results
97(5)
3.7 Concluding Remarks
102(1)
3.8 Summary
103(6)
References
104(5)
4 A Stochastic Framework For Hierarchical System-Level Power Management
109(24)
Peng Rong
Massoud Pedram
4.1 Introduction
109(2)
4.2 Related Work
111(2)
4.3 A Hierarchical DPM Architecture
113(1)
4.4 Modeling
114(8)
4.4.1 Model of the Application Pool
114(4)
4.4.2 Model of the Service Flow Control
118(1)
4.4.3 Model of the Simulated Service Provider
119(1)
4.4.4 Modeling Dependencies between SPs
120(2)
4.5 Policy Optimization
122(3)
4.5.1 Mathematical Formulation
122(1)
4.5.2 Optimal Time-Out Policy for Local Power Manager
123(2)
4.6 Experimental Results
125(5)
4.7 Conclusion
130(3)
References
130(3)
5 Energy-Efficient Reservation Infrastructure For Grids, Clouds, And Networks
133(30)
Anne-Cecile Orgerie
Laurent Lefevre
5.1 Introduction
133(1)
5.2 Related Works
134(4)
5.2.1 Server and Data Center Power Management
135(1)
5.2.2 Node Optimizations
135(1)
5.2.3 Virtualization to Improve Energy Efficiency
136(1)
5.2.4 Energy Awareness in Wired Networking Equipment
136(1)
5.2.5 Synthesis
137(1)
5.3 ERIDIS: Energy-Efficient Reservation Infrastructure for Large-Scale Distributed Systems
138(9)
5.3.1 ERIDIS Architecture
138(3)
5.3.2 Management of the Resource Reservations
141(4)
5.3.3 Resource Management and On/Off Algorithms
145(1)
5.3.4 Energy-Consumption Estimates
146(1)
5.3.5 Prediction Algorithms
146(1)
5.4 EARI: Energy-Aware Reservation Infrastructure for Data Centers and Grids
147(2)
5.4.1 EARI's Architecture
147(1)
5.4.2 Validation of EARI on Experimental Grid Traces
147(2)
5.5 GOC: Green Open Cloud
149(3)
5.5.1 GOC's Resource Manager Architecture
150(2)
5.5.2 Validation of the GOC Framework
152(1)
5.6 HERMES: High Level Energy-Aware Model for Bandwidth Reservation in End-To-End Networks
152(6)
5.6.1 HERMES' Architecture
154(1)
5.6.2 The Reservation Process of HERMES
155(2)
5.6.3 Discussion
157(1)
5.7 Summary
158(5)
References
158(5)
6 Energy-Efficient Job Placement On Clusters, Grids, And Clouds
163(26)
Damien Borgetto
Henri Casanova
Georges Da Costa
Jean-Marc Pierson
6.1 Problem and Motivation
163(1)
6.1.1 Context
163(1)
6.1.2
Chapter Roadmap
164(1)
6.2 Energy-Aware Infrastructures
164(3)
6.2.1 Buildings
165(1)
6.2.2 Context-Aware Buildings
165(1)
6.2.3 Cooling
166(1)
6.3 Current Resource Management Practices
167(3)
6.3.1 Widely Used Resource Management Systems
167(2)
6.3.2 Job Requirement Description
169(1)
6.4 Scientific and Technical Challenges
170(2)
6.4.1 Theoretical Difficulties
170(1)
6.4.2 Technical Difficulties
170(1)
6.4.3 Controlling and Tuning Jobs
171(1)
6.5 Energy-Aware Job Placement Algorithms
172(8)
6.5.1 State of the Art
172(2)
6.5.2 Detailing One Approach
174(6)
6.6 Discussion
180(3)
6.6.1 Open Issues and Opportunities
180(2)
6.6.2 Obstacles for Adoption in Production
182(1)
6.7 Conclusion
183(6)
References
184(5)
7 Comparison And Analysis Of Greedy Energy-Efficient Scheduling Algorithms For Computational Grids
189(26)
Peder Lindberg
James Leingang
Daniel Lysaker
Kashif Bilal
Samee Ullah Khan
Pascal Bouvry
Nasir Ghani
Nasro Min-Allah
Juan Li
7.1 Introduction
189(2)
7.2 Problem Formulation
191(2)
7.2.1 The System Model
191(1)
7.2.1.1 PEs
191(1)
7.2.1.2 DVS
191(1)
7.2.1.3 Tasks
192(1)
7.2.1.4 Preliminaries
192(1)
7.2.2 Formulating the Energy-Makespan Minimization Problem
192(1)
7.3 Proposed Algorithms
193(10)
7.3.1 Greedy Heuristics
194(2)
7.3.1.1 Greedy heuristic scheduling algorithm
196(1)
7.3.1.2 Greedy-min
197(1)
7.3.1.3 Greedy-deadline
198(1)
7.3.1.4 Greedy-max
198(1)
7.3.1.5 MaxMin
199(1)
7.3.1.6 ObFun
199(3)
7.3.1.7 MinMin StdDev
202(1)
7.3.1.8 MinMax StdDev
202(1)
7.4 Simulations, Results, and Discussion
203(8)
7.4.1 Workload
203(1)
7.4.2 Comparative Results
204(1)
7.4.2.1 Small-size problems
204(2)
7.4.2.2 Large-size problems
206(5)
7.5 Related Works
211(1)
7.6 Conclusion
211(4)
References
212(3)
8 Toward Energy-Aware Scheduling Using Machine Learning
215(30)
Josep L.L. Berral
Inigo Goiri
Ramon Nou
Ferran Julia
Josep O. Fito
Jordi Guitart
Ricard Gavalda
Jordi Torres
8.1 Introduction
215(3)
8.1.1 Energetic Impact of the Cloud
216(1)
8.1.2 An Intelligent Way to Manage Data Centers
216(1)
8.1.3 Current Autonomic Computing Techniques
217(1)
8.1.4 Power-Aware Autonomic Computing
217(1)
8.1.5 State of the Art and Case Study
218(1)
8.2 Intelligent Self-Management
218(7)
8.2.1 Classical AI Approaches
219(1)
8.2.1.1 Heuristic algorithms
219(1)
8.2.1.2 AI planning
219(1)
8.2.1.3 Semantic techniques
219(1)
8.2.1.4 Expert systems and genetic algorithms
220(1)
8.2.2 Machine Learning Approaches
220(1)
8.2.2.1 Instance-based learning
221(1)
8.2.2.2 Reinforcement learning
222(3)
8.2.2.3 Feature and example selection
225(1)
8.3 Introducing Power-Aware Approaches
225(5)
8.3.1 Use of Virtualization
226(2)
8.3.2 Turning On and Off Machines
228(1)
8.3.3 Dynamic Voltage and Frequency Scaling
229(1)
8.3.4 Hybrid Nodes and Data Centers
230(1)
8.4 Experiences of Applying ML on Power-Aware Self-Management
230(8)
8.4.1 Case Study Approach
231(1)
8.4.2 Scheduling and Power Trade-Off
231(2)
8.4.3 Experimenting with Power-Aware Techniques
233(3)
8.4.4 Applying Machine Learning
236(2)
8.4.5 Conclusions from the Experiments
238(1)
8.5 Conclusions on Intelligent Power-Aware Self-Management
238(7)
References
240(5)
9 Energy Efficiency Metrics For Data Centers
245(26)
Javid Taheri
Albert Y. Zomaya
9.1 Introduction
245(5)
9.1.1 Background
245(1)
9.1.2 Data Center Energy Use
246(1)
9.1.3 Data Center Characteristics
246(1)
9.1.3.1 Electric power
247(2)
9.1.3.2 Heat removal
249(1)
9.1.4 Energy Efficiency
250(1)
9.2 Fundamentals of Metrics
250(2)
9.2.1 Demand and Constraints on Data Center Operators
250(1)
9.2.2 Metrics
251(1)
9.2.2.1 Criteria for good metrics
251(1)
9.2.2.2 Methodology
252(1)
9.2.2.3 Stability of metrics
252(1)
9.3 Data Center Energy Efficiency
252(8)
9.3.1 Holistic IT Efficiency Metrics
252(2)
9.3.1.1 Fixed versus proportional overheads
254(1)
9.3.1.2 Power versus energy
254(1)
9.3.1.3 Performance versus productivity
255(1)
9.3.2 Code of Conduct
256(1)
9.3.2.1 Environmental statement
256(1)
9.3.2.2 Problem statement
256(1)
9.3.2.3 Scope of the CoC
257(1)
9.3.2.4 Aims and objectives of CoC
258(1)
9.3.3 Power Use in Data Centers
259(1)
9.3.3.1 Data center IT power to utility power relationship
259(1)
9.3.3.2 Chiller efficiency and external temperature
260(1)
9.4 Available Metrics
260(7)
9.4.1 The Green Grid
261(1)
9.4.1.1 Power usage effectiveness (PUE)
261(1)
9.4.1.2 Data center efficiency (DCE)
262(1)
9.4.1.3 Data center infrastructure efficiency (DCiE)
262(1)
9.4.1.4 Data center productivity (DCP)
263(1)
9.4.2 McKinsey
263(1)
9.4.3 Uptime Institute
264(1)
9.4.3.1 Site infrastructure power overhead multiplier (SI-POM)
265(1)
9.4.3.2 IT hardware power overhead multiplier (H-POM)
266(1)
9.4.3.3 DC hardware compute load per unit of computing work done
266(1)
9.4.3.4 Deployed hardware utilization ratio (DH-UR)
266(1)
9.4.3.5 Deployed hardware utilization efficiency (DH-UE)
267(1)
9.5 Harmonizing Global Metrics for Data Center Energy Efficiency
267(4)
References
268(3)
10 Autonomic Green Computing In Large-Scale Data Centers
271(30)
Haoting Luo
Bithika Khargharia
Salim Hariri
Youssif Al-Nashif
10.1 Introduction
271(1)
10.2 Related Technologies and Techniques
272(11)
10.2.1 Power Optimization Techniques in Data Centers
272(1)
10.2.2 Design Model
273(1)
10.2.3 Networks
274(1)
10.2.4 Data Center Power Distribution
275(1)
10.2.5 Data Center Power-Efficient Metrics
276(1)
10.2.6 Modeling Prototype and Testbed
277(1)
10.2.7 Green Computing
278(2)
10.2.8 Energy Proportional Computing
280(1)
10.2.9 Hardware Virtualization Technology
281(1)
10.2.10 Autonomic Computing
282(1)
10.3 Autonomic Green Computing: A Case Study
283(14)
10.3.1 Autonomic Management Platform
285(1)
10.3.1.1 Platform architecture
285(1)
10.3.1.2 DEVS-based modeling and simulation platform
285(2)
10.3.1.3 Workload generator
287(1)
10.3.2 Model Parameter Evaluation
288(1)
10.3.2.1 State transitioning overhead
288(1)
10.3.2.2 VM template evaluation
289(2)
10.3.2.3 Scalability analysis
291(1)
10.3.3 Autonomic Power Efficiency Management Algorithm (Performance Per Watt)
291(2)
10.3.4 Simulation Results and Evaluation
293(3)
10.3.4.1 Analysis of energy and performance trade-offs
296(1)
10.4 Conclusion and Future Directions
297(4)
References
298(3)
11 Energy And Thermal Aware Scheduling In Data Centers
301(38)
Gaurav Dhiman
Raid Ayoub
Tajana S. Rosing
11.1 Introduction
301(1)
11.2 Related Work
302(3)
11.3 Intermachine Scheduling
305(10)
11.3.1 Performance and Power Profile of VMs
305(4)
11.3.2 Architecture
309(1)
11.3.2.1 vgnode
309(1)
11.3.2.2 vgxen
310(2)
11.3.2.3 vgdom
312(1)
11.3.2.4 vgserv
312(3)
11.4 Intramachine Scheduling
315(6)
11.4.1 Air-Forced Thermal Modeling and Cost
316(1)
11.4.2 Cooling Aware Dynamic Workload Scheduling
317(1)
11.4.3 Scheduling Mechanism
318(1)
11.4.4 Cooling Costs Predictor
319(2)
11.5 Evaluation
321(12)
11.5.1 Intermachine Scheduler (vGreen)
321(2)
11.5.2 Heterogeneous Workloads
323(2)
11.5.2.1 Comparison with DVFS policies
325(3)
11.5.2.2 Homogeneous workloads
328(1)
11.5.3 Intramachine Scheduler (Cool and Save)
328(3)
11.5.3.1 Results
331(2)
11.5.3.2 Overhead of CAS
333(1)
11.6 Conclusion
333(6)
References
334(5)
12 QOS-Aware Power Management In Data Centers
339(22)
Jiayu Gong
Cheng-Zhong Xu
12.1 Introduction
339(1)
12.2 Problem Classification
340(4)
12.2.1 Objective and Constraint
340(1)
12.2.2 Scope and Time Granularities
340(1)
12.2.3 Methodology
341(1)
12.2.4 Power Management Mechanism
342(2)
12.3 Energy Efficiency
344(7)
12.3.1 Energy-Efficiency Metrics
344(2)
12.3.2 Improving Energy Efficiency
346(1)
12.3.2.1 Energy minimization with performance guarantee
346(2)
12.3.2.2 Performance maximization under power budget
348(1)
12.3.2.3 Trade-off between power and performance
348(2)
12.3.3 Energy-Proportional Computing
350(1)
12.4 Power Capping
351(2)
12.5 Conclusion
353(8)
References
356(5)
13 Energy-Efficient Storage Systems For Data Centers
361(16)
Sudhanva Gurumurthi
Anand Sivasubramaniam
13.1 Introduction
361(1)
13.2 Disk Drive Operation and Disk Power
362(4)
13.2.1 An Overview of Disk Drives
362(1)
13.2.2 Sources of Disk Power Consumption
363(2)
13.2.3 Disk Activity and Power Consumption
365(1)
13.3 Disk and Storage Power Reduction Techniques
366(5)
13.3.1 Exploiting the STANDBY State
368(1)
13.3.2 Reducing Seek Activity
369(1)
13.3.3 Achieving Energy Proportionality
369(1)
13.3.3.1 Hardware approaches
369(1)
13.3.3.2 Software approaches
370(1)
13.4 Using Nonvolatile Memory and Solid-State Disks
371(1)
13.5 Conclusions
372(5)
References
373(4)
14 Autonomic Energy/Performance Optimizations For Memory In Servers
377(18)
Bithika Khargharia
Mazin Yousif
14.1 Introduction
378(2)
14.2 Classifications of Dynamic Power Management Techniques
380(2)
14.2.1 Heuristic and Predictive Techniques
380(1)
14.2.2 QoS and Energy Trade-Offs
381(1)
14.3 Applications of Dynamic Power Management (DPM)
382(2)
14.3.1 Power Management of System Components in Isolation
382(1)
14.3.2 Joint Power Management of System Components
383(1)
14.3.3 Holistic System-Level Power Management
383(1)
14.4 Autonomic Power and Performance Optimization of Memory Subsystems in Server Platforms
384(7)
14.4.1 Adaptive Memory Interleaving Technique for Power and Performance Management
384(2)
14.4.1.1 Formulating the optimization problem
386(3)
14.4.1.2 Memory appflow
389(1)
14.4.2 Industry Techniques
389(1)
14.4.2.1 Enhancements in memory hardware design
390(1)
14.4.2.2 Adding more operating states
390(1)
14.4.2.3 Faster transition to and from low power states
390(1)
14.4.2.4 Memory consolidation
390(1)
14.5 Conclusion
391(4)
References
391(4)
15 ROD: A Practical Approach To Improving Reliability Of Energy-Efficient Parallel Disk Systems
395(22)
Shu Yin
Xiaojun Ruan
Adam Manzanares
Xiao Qin
15.1 Introduction
395(1)
15.2 Modeling Reliability of Energy-Efficient Parallel Disks
396(5)
15.2.1 The MINT Model
396(2)
15.2.1.1 Disk utilization
398(1)
15.2.1.2 Temperature
398(1)
15.2.1.3 Power-state transition frequency
399(1)
15.2.1.4 Single disk reliability model
399(1)
15.2.2 MAID, Massive Arrays of Idle Disks
400(1)
15.3 Improving Reliability of MAID via Disk Swapping
401(4)
15.3.1 Improving Reliability of Cache Disks in MAID
401(3)
15.3.2 Swapping Disks Multiple Times
404(1)
15.4 Experimental Results and Evaluation
405(6)
15.4.1 Experimental Setup
405(1)
15.4.2 Disk Utilization
406(1)
15.4.3 The Single Disk Swapping Strategy
406(3)
15.4.4 The Multiple Disk Swapping Strategy
409(2)
15.5 Related Work
411(1)
15.6 Conclusions
412(5)
References
413(4)
16 Embracing The Memory And I/O Walls For Energy-Efficient Scientific Computing
417(26)
Chung-Hsing Hsu
Wu-Chun Feng
16.1 Introduction
417(3)
16.2 Background and Related Work
420(3)
16.2.1 DVFS-Enabled Processors
420(1)
16.2.2 DVFS Scheduling Algorithms
421(1)
16.2.3 Memory-Aware, Interval-Based Algorithms
422(1)
16.3 β-Adaptation: A New DVFS Algorithm
423(6)
16.3.1 The Compute-Boundedness Metric, β
423(1)
16.3.2 The Frequency Calculating Formula, f*
424(1)
16.3.3 The Online β Estimation
425(2)
16.3.4 Putting It All Together
427(2)
16.4 Algorithm Effectiveness
429(9)
16.4.1 A Comparison to Other DVFS Algorithms
429(3)
16.4.2 Frequency Emulation
432(4)
16.4.3 The Minimum Dependence to the PMU
436(2)
16.5 Conclusions and Future Work
438(5)
References
439(4)
17 Multiple Frequency Selection In Dvfs-Enabled Processors To Minimize Energy Consumption
443(22)
Nikzad Babaii Rizvandi
Albert Y. Zomaya
Young Choon Lee
Ali Javadzadeh Boloori
Javid Taheri
17.1 Introduction
443(1)
17.2 Energy Efficiency in HPC Systems
444(2)
17.3 Exploitation of Dynamic Voltage-Frequency Scaling
446(2)
17.3.1 Independent Slack Reclamation
446(1)
17.3.2 Integrated Schedule Generation
447(1)
17.4 Preliminaries
448(2)
17.4.1 System and Application Models
448(1)
17.4.2 Energy Model
448(2)
17.5 Energy-Aware Scheduling via DVFS
450(6)
17.5.1 Optimum Continuous Frequency
450(1)
17.5.2 Reference Dynamic Voltage-Frequency Scaling (RDVFS)
451(1)
17.5.3 Maximum-Minimum-Frequency for Dynamic Voltage-Frequency Scaling (MMF-DVFS)
452(1)
17.5.4 Multiple Frequency Selection for Dynamic Voltage-Frequency Scaling (MFS-DVFS)
453(1)
17.5.4.1 Task eligibility
454(2)
17.6 Experimental Results
456(5)
17.6.1 Simulation Settings
456(2)
17.6.2 Results
458(3)
17.7 Conclusion
461(4)
References
461(4)
18 The Paramountcy Of Reconfigurable Computing
465(84)
Reiner Hartenstein
18.1 Introduction
465(1)
18.2 Why Computers are Important
466(6)
18.2.1 Computing for a Sustainable Environment
470(2)
18.3 Performance Progress Stalled
472(16)
18.3.1 Unaffordable Energy Consumption of Computing
473(2)
18.3.2 Crashing into the Programming Wall
475(13)
18.4 The Tail is Wagging the Dog (Accelerators)
488(6)
18.4.1 Hardwired Accelerators
489(1)
18.4.2 Programmable Accelerators
490(4)
18.5 Reconfigurable Computing
494(32)
18.5.1 Speedup Factors by FPGAs
498(3)
18.5.2 The Reconfigurable Computing Paradox
501(4)
18.5.3 Saving Energy by Reconfigurable Computing
505(1)
18.5.3.1 Traditional green computing
506(1)
18.5.3.2 The role of graphics processors
507(1)
18.5.3.3 Wintel versus ARM
508(3)
18.5.4 Reconfigurable Computing is the Silver Bullet
511(1)
18.5.4.1 A new world model of computing
511(3)
18.5.5 The Twin-Paradigm Approach to Tear Down the Wall
514(3)
18.5.6 A Mass Movement Needed as Soon as Possible
517(1)
18.5.6.1 Legacy software from the mainframe age
518(1)
18.5.7 How to Reinvent Computing
519(7)
18.6 Conclusions
526(23)
References
529(20)
19 Workload Clustering For Increasing Energy Savings On Embedded Mpsocs
549(18)
Ozcan Ozturk
Mahmut Kandemir
Sri Hari Krishna Narayanan
19.1 Introduction
549(1)
19.2 Embedded MPSoC Architecture, Execution Model, and Related Work
550(1)
19.3 Our Approach
551(9)
19.3.1 Overview
551(2)
19.3.2 Technical Details and Problem Formulation
553(1)
19.3.2.1 System and job model
553(1)
19.3.2.2 Mathematical programing model
554(3)
19.3.2.3 Example
557(3)
19.4 Experimental Evaluation
560(4)
19.5 Conclusions
564(3)
References
565(2)
20 Energy-Efficient Internet Infrastructure
567(26)
Weirong Jiang
Viktor K. Prasanna
20.1 Introduction
567(4)
20.1.1 Performance Challenges
568(2)
20.1.2 Existing Packet Forwarding Approaches
570(1)
20.1.2.1 Software approaches
570(1)
20.1.2.2 Hardware approaches
571(1)
20.2 SRAM-Based Pipelined IP Lookup Architectures: Alternative to TCAMs
571(2)
20.3 Data Structure Optimization for Power Efficiency
573(7)
20.3.1 Problem Formulation
574(1)
20.3.1.1 Non-pipelined and pipelined engines
574(1)
20.3.1.2 Power function of SRAM
575(1)
20.3.2 Special Case: Uniform Stride
576(1)
20.3.3 Dynamic Programming
576(1)
20.3.4 Performance Evaluation
577(1)
20.3.4.1 Results for non-pipelined architecture
578(1)
20.3.4.2 Results for pipelined architecture
578(2)
20.4 Architectural Optimization to Reduce Dynamic Power Dissipation
580(8)
20.4.1 Analysis and Motivation
581(1)
20.4.1.1 Traffic locality
582(1)
20.4.1.2 Traffic rate variation
582(1)
20.4.1.3 Access frequency on different stages
583(1)
20.4.2 Architecture-Specific Techniques
583(1)
20.4.2.1 Inherent caching
584(1)
20.4.2.2 Local clocking
584(1)
20.4.2.3 Fine-grained memory enabling
585(1)
20.4.3 Performance Evaluation
585(3)
20.5 Related Work
588(1)
20.6 Summary
589(4)
References
589(4)
21 Demand Response In The Smart Grid: A Distributed Computing Perspective
593(22)
Chen Wang
Martin De Groot
21.1 Introduction
593(2)
21.2 Demand Response
595(5)
21.2.1 Existing Demand Response Programs
595(2)
21.2.2 Demand Response Supported by the Smart Grid
597(3)
21.3 Demand Response as a Distributed System
600(11)
21.3.1 An Overlay Network for Demand Response
600(2)
21.3.2 Event Driven Demand Response
602(2)
21.3.3 Cost Driven Demand Response
604(5)
21.3.4 A Decentralized Demand Response Framework
609(1)
21.3.5 Accountability of Coordination Decision Making
610(1)
21.4 Summary
611(4)
References
611(4)
22 Resource Management For Distributed Mobile Computing
615(38)
Jong-Kook Kim
22.1 Introduction
615(2)
22.2 Single-Hop Energy-Constrained Environment
617(18)
22.2.1 System Model
617(3)
22.2.2 Related Work
620(1)
22.2.3 Heuristic Descriptions
621(1)
22.2.3.1 Mapping event
621(1)
22.2.3.2 Scheduling communications
621(1)
22.2.3.3 Opportunistic load balancing and minimum energy greedy heuristics
622(1)
22.2.3.4 ME-MC heuristic
622(2)
22.2.3.5 ME-ME heuristic
624(1)
22.2.3.6 CRME heuristic
625(1)
22.2.3.7 Originator and random
626(1)
22.2.3.8 Upper bound
626(2)
22.2.4 Simulation Model
628(2)
22.2.5 Results
630(4)
22.2.6 Summary
634(1)
22.3 Multihop Distributed Mobile Computing Environment
635(12)
22.3.1 The Multihop System Model
635(1)
22.3.2 Energy-Aware Routing Protocol
636(1)
22.3.2.1 Overview
636(1)
22.3.2.2 DSDV
637(1)
22.3.2.3 DSDV remaining energy
637(1)
22.3.2.4 DSDV-energy consumption per remaining energy
637(1)
22.3.3 Heuristic Description
638(1)
22.3.3.1 Random
638(1)
22.3.3.2 Estimated minimum total energy (EMTE)
638(1)
22.3.3.3 K-percent-speed (KPS) and K-percent-energy (KPE)
639(1)
22.3.3.4 Energy ratio and distance (ERD)
639(1)
22.3.3.5 ETC and distance (ETCD)
640(1)
22.3.3.6 Minimum execution time (MET)
640(1)
22.3.3.7 Minimum completion time (MCT) and minimum completion time with DVS (MCT-DVS)
640(1)
22.3.3.8 Switching algorithm (SA)
640(1)
22.3.4 Simulation Model
641(2)
22.3.5 Results
643(1)
22.3.5.1 Distributed resource management
643(1)
22.3.5.2 Energy-aware protocol
644(1)
22.3.6 Summary
644(3)
22.4 Future Work
647(6)
References
647(6)
23 An Energy-Aware Framework For Mobile Data Mining
653(20)
Carmela Comito
Domenico Talia
Paolo Trunfio
23.1 Introduction
653(1)
23.2 System Architecture
654(3)
23.3 Mobile Device Components
657(2)
23.4 Energy Model
659(5)
23.5 Clustering Scheme
664(6)
23.5.1 Clustering the M2M Architecture
666(4)
23.6 Conclusion
670(3)
References
670(3)
24 Energy Awareness And Efficiency In Wireless Sensor Networks: From Physical Devices To The Communication Link
673(36)
Flavia C. Delicato
Paulo F. Pires
24.1 Introduction
673(3)
24.2 WSN and Power Dissipation Models
676(7)
24.2.1 Network and Node Architecture
676(3)
24.2.2 Sources of Power Dissipation in WSNs
679(4)
24.3 Strategies for Energy Optimization
683(18)
24.3.1 Intranode Level
684(1)
24.3.1.1 Duty cycling
685(6)
24.3.1.2 Adaptive sensing
691(2)
24.3.1.3 Dynamic voltage scale (DVS)
693(1)
24.3.1.4 OS task scheduling
694(1)
24.3.2 Internode Level
695(1)
24.3.2.1 Transmission power control
695(1)
24.3.2.2 Dynamic modulation scaling
696(2)
24.3.2.3 Link layer optimizations
698(3)
24.4 Final Remarks
701(8)
References
702(7)
25 Network-Wide Strategies For Energy Efficiency In Wireless Sensor Networks
709(42)
Flavia C. Delicato
Paulo F. Pires
25.1 Introduction
709(2)
25.2 Data Link Layer
711(8)
25.2.1 Topology Control Protocols
712(2)
25.2.2 Energy-Efficient MAC Protocols
714(2)
25.2.2.1 Scheduled MAC protocols in WSNs
716(1)
25.2.2.2 Contention-based MAC protocols
717(2)
25.3 Network Layer
719(6)
25.3.1 Flat and Hierarchical Protocols
722(3)
25.4 Transport Layer
725(4)
25.5 Application Layer
729(11)
25.5.1 Task Scheduling
729(4)
25.5.2 Data Aggregation and Data Fusion in WSNs
733(2)
25.5.2.1 Approaches of data fusion for energy efficiency
735(1)
25.5.2.2 Data aggregation strategies
736(4)
25.6 Final Remarks
740(11)
References
741(10)
26 Energy Management In Heterogeneous Wireless Health Care Networks
751(36)
Nima Nikzad
Priti Aghera
Piero Zappi
Tajana S. Rosing
26.1 Introduction
751(2)
26.2 System Model
753(2)
26.2.1 Health Monitoring Task Model
753(2)
26.3 Collaborative Distributed Environmental Sensing
755(5)
26.3.1 Node Neighborhood and Localization Rate
757(1)
26.3.2 Energy Ratio and Sensing Rate
758(1)
26.3.3 Duty Cycling and Prediction
759(1)
26.4 Task Assignment in a Body Area Network
760(11)
26.4.1 Optimal Task Assignment
760(2)
26.4.2 Dynamic Task Assignment
762(1)
26.4.2.1 DynAGreen algorithm
763(5)
26.4.2.2 DynAGreenLife algorithm
768(3)
26.5 Results
771(13)
26.5.1 Collaborative Sensing
771(1)
26.5.1.1 Results
772(4)
26.5.2 Dynamic Task Assignment
776(1)
26.5.2.1 Performance in static conditions
777(3)
26.5.2.2 Dynamic adaptability
780(4)
26.6 Conclusion
784(3)
References
785(2)
Index 787
ALBERT Y. ZOMAYA is the Chair Professor of High Performance Computing & Networking in the School of Information Technologies, The University of Sydney. He is a Fellow of the IEEE, the American Association for the Advancement of Science, and the Institution of Engineering and Technology, and a Distinguished Engineer of the ACM. He has authored seven books and some 400 articles in technical journals.

YOUNG CHOON LEE, PhD, is with the Centre for Distributed and High Performance Computing, School of Information Technologies, The University of Sydney.
Lisainfo e-raamatute kohta Lisainfo e-raamatute kohta
Püsilink: https://www.kriso.ee/db/97811183419882e.html
Märksõnad: