Energy-Efficient Distributed Computing Systems

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