| List of Figures | p. xi |
| List of Tables | p. xv |
| Contributing Authors | p. xvii |
| Preface | p. xix |
| Low Power Operating System for Heterogeneous Wireless Communication System | p. 1 |
| Introduction | p. 2 |
| Event-driven versus General-purpose OS | p. 3 |
| PicoRadio II Protocol Design | p. 3 |
| General-purpose Multi-tasking OS | p. 4 |
| Event-driven OS | p. 8 |
| Comparison Summary | p. 9 |
| Low Power Reactive OS for Heterogeneous Architectures | p. 12 |
| Event-driven Global Scheduler and Power Management | p. 12 |
| TinyOS Limitations and Proposed Extensions | p. 14 |
| Conclusion and Future Work | p. 15 |
| References | p. 16 |
| A Modified Dual-Priority Scheduling Algorithm for Hard Real-Time Systems to Improve Energy Savings | p. 17 |
| Introduction | p. 17 |
| Dual-Priority Scheduling | p. 19 |
| Power-Low Modified Dual-Priority Scheduling | p. 21 |
| Experimental Results | p. 28 |
| Summary | p. 36 |
| References | p. 36 |
| Toward the Placement of Power Management Points in Real-Time Applications | p. 37 |
| Introduction | p. 37 |
| Model | p. 39 |
| Sources of Overhead | p. 40 |
| Computing the New Speed | p. 40 |
| Setting the New Speed | p. 40 |
| Speed Adjustment Schemes | p. 41 |
| Proportional Dynamic Power Management | p. 41 |
| Dynamic Greedy Power Management | p. 42 |
| Evaluation of Power Management Schemes | p. 43 |
| Optimal Number of PMPs | p. 44 |
| Evaluation of the Analytical Model | p. 45 |
| Conclusion | p. 48 |
| Derivation of Formulas | p. 48 |
| References | p. 51 |
| Energy Characterization of Embedded Real-Time Operating Systems | p. 53 |
| Introduction | p. 53 |
| Related Work | p. 55 |
| System Overview | p. 56 |
| The Hardware Platform | p. 56 |
| RTOS overview | p. 57 |
| Characterization Strategy | p. 59 |
| RTOS Characterization Results | p. 60 |
| Kernel Services | p. 60 |
| I/O Drivers | p. 62 |
| Burstiness Test | p. 62 |
| Clock Speed Test | p. 63 |
| Resource Contention Test | p. 64 |
| Application Example: RTOS vs Stand-alone | p. 65 |
| Cache Related Effects in Thread Switching | p. 66 |
| Summary of Findings | p. 66 |
| Conclusions | p. 67 |
| References | p. 72 |
| Dynamic Cluster Reconfiguration for Power and Performance | p. 75 |
| Motivation | p. 77 |
| Cluster Configuration and Load Distribution | p. 78 |
| Overview | p. 78 |
| Implementations | p. 81 |
| Methodology | p. 83 |
| Experimental Results | p. 84 |
| Related Work | p. 89 |
| Conclusions | p. 91 |
| References | p. 91 |
| Energy Management of Virtual Memory on Diskless Devices | p. 95 |
| Introduction | p. 96 |
| Related Work | p. 97 |
| Problem Formulation | p. 98 |
| EEL[subscript RM] Prototype Compiler | p. 100 |
| Phase 1 - Analysis | p. 100 |
| Phase 2 - Code Generation | p. 101 |
| Performance Model | p. 102 |
| Example | p. 102 |
| Implementation Issues | p. 103 |
| Experiments | p. 105 |
| Benchmark Characteristics | p. 106 |
| Simulation Results | p. 107 |
| Future Work | p. 110 |
| Conclusion | p. 111 |
| References | p. 111 |
| Propagating Constants Past Software to Hardware Peripherals on Fixed-Application Embedded Systems | p. 115 |
| Introduction | p. 116 |
| Example | p. 119 |
| Parameters in Cores | p. 120 |
| Propagating Constants from Software to Hardware | p. 123 |
| Experiments | p. 125 |
| 8255A Programmable Peripheral Interface | p. 126 |
| 8237A DMA Controller | p. 127 |
| PC16550A UART | p. 128 |
| Free-DCT-L Core | p. 128 |
| Results | p. 131 |
| Future Work | p. 133 |
| Conclusions | p. 134 |
| References | p. 134 |
| Constructive Timing Violation for Improving Energy Efficiency | p. 137 |
| Introduction | p. 137 |
| Low Power via Fault-Tolerance | p. 139 |
| Evaluation Methodology | p. 143 |
| Simulation Results | p. 143 |
| Related Work | p. 147 |
| Conclusion and Future Work | p. 151 |
| References | p. 151 |
| Power Modeling and Reduction of VLIW Processors | p. 155 |
| Introduction | p. 155 |
| Cycle-Accurate VLIW Power Simulation | p. 156 |
| IMPACT Architecture Framework | p. 156 |
| Power Models | p. 157 |
| PowerImpact | p. 158 |
| Clock Ramping | p. 159 |
| Clock Ramping with Hardware Prescan (CRHP) | p. 160 |
| Clock Ramping with Compiler-based Prediction (CRCP) | p. 162 |
| Basic CRCP Algorithm | p. 162 |
| Reduction of Redundant Ramp-up Instructions | p. 164 |
| Control Flow | p. 165 |
| Load Instructions | p. 165 |
| Experimental Results | p. 165 |
| Conclusions and Discussion | p. 169 |
| References | p. 170 |
| Low-Power Design of Turbo Decoder with Exploration of Energy-Throughput Trade-off | p. 173 |
| Introduction | p. 173 |
| Data Transfer and Storage Exploration Methodology | p. 176 |
| Global Data Flow and Loop Transformations | p. 178 |
| Removal of Interleaver Memory | p. 178 |
| Enabling Parallelism | p. 179 |
| Storage Cycle Budget Distribution | p. 180 |
| Memory Hierarchy Layer Assignment | p. 181 |
| Data Restructuring | p. 182 |
| Loop Transformations for Parallelization | p. 183 |
| Loop Merging | p. 183 |
| Loop Pipelining | p. 184 |
| Partial Loop Unrolling | p. 184 |
| Loop Transformation Results | p. 185 |
| Storage Bandwidth Optimization | p. 185 |
| Memory Organization | p. 186 |
| Memory Organization Exploration | p. 186 |
| Memory Organization Decision | p. 188 |
| Conclusions | p. 190 |
| References | p. 190 |
| Static Analysis of Parameterized Loop Nests for Energy Efficient Use of Data Caches | p. 193 |
| Introduction | p. 193 |
| Energy and Line Size | p. 195 |
| Background | p. 195 |
| The Parameterized Loop Analysis | p. 197 |
| Reduction to Single Reference Interference | p. 199 |
| Interference and Reuse Trade-off | p. 200 |
| STAMINA Implementation Results | p. 200 |
| Swim from SPEC 2000 | p. 201 |
| Self Interference | p. 201 |
| Tiling and Matrix Multiply | p. 202 |
| Summary and Future Work | p. 203 |
| References | p. 203 |
| A Fresh Look at Low-Power Mobile Computing | p. 209 |
| Introduction | p. 209 |
| Architecture | p. 211 |
| Handover and the Quantization of Computational Resources | p. 212 |
| Standardization of Execution Environment's Parameters | p. 214 |
| A Commercial Vision: Impact on Billing, Customer Loyalty and Churn | p. 215 |
| Segmentation of Functionality: The XU-MS Split | p. 215 |
| Use of Field-Programmable Hardware in the Mobile Station | p. 217 |
| Special End-To-End Application Requirements | p. 217 |
| Status and Research Vision | p. 218 |
| References | p. 219 |
| Index | p. 221 |
| Table of Contents provided by Ingram. All Rights Reserved. |