Dependable Embedded Systems.

• Superior document: Embedded Systems Series
• Author: Henkel, Jörg.
• TeilnehmendeR: Dutt, Nikil.
• Place / Publishing House: Cham : : Springer International Publishing AG,, 2020.; {copy}2021.
• Year of Publication: 2020
• Edition: 1st ed.
• Language: English
• Series: Embedded Systems Series
• Physical Description: 1 online resource (606 pages)

Table of Contents

Intro -- Preface -- Fabrication and Design-Time Effects -- Yield and Process Variations -- Complexity -- Operation and Run-Time Effects -- Aging Effects -- Thermal Effects -- Soft Errors -- Contents -- RAP Model-Enabling Cross-Layer Analysis and Optimization for System-on-Chip Resilience -- 1 Introduction/Motivation -- 2 Resilience Articulation Point (RAP) Basics -- 3 Related Work -- 4 Fault Abstraction at Lower Levels -- 4.1 SRAM Errors -- 4.1.1 SRAM Errors due to Particle Strikes (Qcrit) -- 4.1.2 SRAM Errors due to Noise (SVNM) -- 4.1.3 SRAM Errors Due to Read/Write Failures (Read Delay/WTV) -- 4.1.4 SRAM Errors due to Supply Voltage Drop -- 5 Architecture Level Analysis and Countermeasures -- 5.1 Instruction Vulnerability -- 5.2 Data Vulnerability Analysis and Mitigation -- 5.3 Dynamic Testing -- 6 Application-Level Optimization-Autonomous Robot -- References -- Part I Cross-Layer from Operating System to Application -- Soft Error Handling for Embedded Systems using Compiler-OS Interaction -- 1 New Requirements for Fault Tolerance -- 2 Semantics of Errors -- 3 FEHLER System Overview and Semantic Annotations -- 4 Timing Behavior -- 5 Static Analyses -- 6 FEHLER Runtime System -- 7 Use Case: A Fault-Tolerant QoS-Aware Soft Real-time Application -- 8 Use Case: Adaptive Error Handling in Control Applications -- 9 Application of FEHLER to Approximate Computing -- 10 Summary and Outlook -- References -- ASTEROID and the Replica-Aware Co-schedulingfor Mixed-Criticality -- 1 The ASTEROID Project -- 1.1 Motivation -- 1.2 Overview -- 2 Replica-Aware Co-scheduling for Mixed-Criticality -- 2.1 Motivation -- 2.2 Related Work -- 3 System, Task, and Error Models -- 3.1 System Model -- 3.2 Task Model -- 3.3 Error Model -- 3.4 Offsets -- 4 Response-Time Analysis -- 4.1 Fork-Join Tasks -- 4.2 Independent Tasks -- 4.3 Error Recovery -- 5 Experimental Evaluation.

5.1 Evaluation with Benchmark Applications -- 5.1.1 Characterization -- 5.1.2 Evaluation of Fork-Join Tasks -- 5.1.3 Evaluation of Independent Tasks -- 5.2 Evaluation with Synthetic Workload -- 5.2.1 Evaluation of Fork-Join Tasks -- 5.2.2 Evaluation of Independent Tasks -- 6 Conclusion -- References -- Dependability Aspects in Configurable Embedded OperatingSystems -- 1 Introduction -- 2 Related Work -- 3 dOSEK: A Dependable RTOS for Automotive Applications -- 3.1 Development of a Fault-Avoiding Operating System -- 3.2 Implementing a Fault-Detecting Operating System -- 3.3 Evaluation -- 3.3.1 Fault-Injection Results -- 3.3.2 Memory- and Runtime Costs -- 4 Modularizing Software-Based Memory Error Detection and Correction -- 4.1 Generic Object Protection with AspectC++ -- 4.1.1 Generic Introductions by Compile-Time Introspection -- 4.1.2 Advice for Control Flow and Data Access -- 4.2 Implementation and Evaluation -- 4.2.1 EDM/ERM Variants -- 4.2.2 Evaluation Setup -- 4.2.3 Optimizing the Generic Object Protection -- 4.2.4 Protection Effectiveness and Overhead -- 4.3 Discussion -- 5 Conserving Consistent State in Persistent Memory with Software Transactional Memory -- 5.1 System Model -- 5.2 Concepts of DNV Memory -- 5.3 Evaluation -- 5.4 Discussion -- 6 Summary -- References -- Part II Cross-Layer Dependability: From Architecture to Software and Operating System -- Increasing Reliability Using Adaptive Cross-Layer Techniques in DRPs: Just-Safe-Enough Responses to Reliability Threats -- 1 Introduction -- 2 Dynamically Reconfigurable Processors -- 3 Exploiting Architectural Redundancy for Increased Reliability -- 3.1 Realizing Low-Cost TMR Using PE Clusters -- 3.2 DRPs as Redundancy for CPU Pipelines -- 3.3 Dynamic Testing -- 3.4 Dynamic Remapping -- 3.5 Testing Reliability Schemes in Hardware -- 4 Device-Level State and Countermeasures.

5 Synergistic Effects of Cross-Layer Approaches -- 6 Conclusion -- References -- Dependable Software Generation and Execution on EmbeddedSystems -- 1 Overview -- 2 Dependability Modeling and Estimation -- 3 Dependability-Driven Compilation -- 3.1 Dependability-Driven Software Transformations -- 3.2 Dependability-Driven Instruction Scheduling -- 3.3 Dependability-Driven Selective Instruction Redundancy -- 4 Dependability-Driven System Software -- 4.1 Joint Consideration of Functional and Timing Dependability -- 4.2 Adaptive Dependability Tuning in Multi-Core Systems -- 5 Resilient Design for System Software -- 5.1 Adaptive Soft Error Handling -- 5.2 Dynamic Real-Time Guarantees -- 5.3 Probabilistic Deadline-Miss Analyses -- 6 Application-Specific Dependability -- 7 Conclusion -- References -- Fault-Tolerant Computing with Heterogeneous Hardening Modes -- 1 Introduction -- 2 Fault-Tolerant Heterogeneous Processors -- 2.1 Hardening Embedded Processors -- 2.2 Reliability Techniques for Multi-Level Caches -- 2.2.1 Improving the Reliability of Last-Level Caches -- 2.2.2 Improving the Reliability of the Complete Cache Hierarchy -- 3 Heterogeneous Reliability Modes of Out-of-Order Superscalar Cores -- 3.1 Experimental Setup -- 3.2 Vulnerability Analysis of Out-of-Order Superscalar Processors -- 3.3 Methodology for Hardening Out-of-Order Superscalar Processors -- 3.3.1 Full-Processor Vulnerability Factor -- 3.3.2 Heterogeneous Reliability Modes for ALPHA Cores -- 3.3.3 State Compression Techniques -- 4 Run-Time Systems for Heterogeneous Fault-Tolerance -- 5 Conclusion -- References -- Thermal Management and Communication Virtualization for Reliability Optimization in MPSoCs -- 1 Overview -- 2 Impact of Temperature on Reliability -- 3 Temperature Estimation via Simulation or Measurement -- 3.1 Thermal Simulation -- 3.2 Thermal Measurement.

4 Modeling Impact of Temperature at System Level -- 4.1 Figures of Merit -- 4.2 Direct Impact of Temperature -- 4.3 Aging as Indirect Impact of Temperature -- 5 System-Level Management -- 5.1 Voltage Scaling -- 5.2 Task Migration -- 6 Architecture Support -- 6.1 NoC Virtualization -- 6.2 Advanced Communication Reconfiguration Using Protection Switching -- 6.3 Adaptive Modular Redundancy (AMR) -- 7 Cross-Layer -- 8 Conclusion -- References -- Lightweight Software-Defined Error Correction for Memories -- 1 Software-Defined Error Correcting Codes (SDECC) -- 1.1 SDECC Theory -- 1.1.1 Computing the List of Candidates -- 1.1.2 SDECC Analysis of Existing ECCs -- 1.2 SDECC Architecture -- 1.3 Data Recovery Policy -- 1.3.1 Observations on Data Similarity -- 1.4 Reliability Evaluation -- 1.4.1 Methodology -- 1.4.2 Recovery Breakdown -- 2 Software-Defined Error-Localizing Codes (SDELC): Lightweight Recovery from Soft Faults at Runtime -- 2.1 Ultra-Lightweight Error-Localizing Codes (UL-ELC) -- 2.2 Recovering SEUs in Instruction Memory -- 2.3 Recovering SEUs in Data Memory -- 2.4 SDELC Architecture -- 2.5 Soft Fault Recovery Using SDELC -- 2.5.1 Overall Results -- 3 Parity++ : Lightweight Error Correction for Last Level Caches and Embedded Memories -- 3.1 Application Characteristics -- 3.2 Parity++ Theory -- 3.3 Error Detection and Correction -- 3.4 Architecture -- 3.5 Experimental Methodology -- 3.6 Results and Discussion -- References -- Resource Management for Improving Overall Reliability of Multi-Processor Systems-on-Chip -- 1 Introduction -- 1.1 Background -- 1.2 Related Work -- 1.3 Soft-Error Reliability Model -- 1.4 Lifetime Reliability Model -- 2 LTR and SER Optimization -- 2.1 LTR Optimization -- 2.2 SER Optimization -- 3 Trade-Off Between LTR and SER -- 3.1 ``Big-Little'' MPSoCs -- 3.2 CPU-GPU Integrated MPSoCs -- 4 Conclusion -- References.

Part III Cross-Layer Resilience: Bridging the Gap Between Circuit and Architectural Layer -- Cross-Layer Resilience Against Soft Errors: Key Insights -- 1 Introduction -- 2 Evaluation of Soft Error Resilience Using Fault Injection -- 2.1 Overview on Fault Injection Methods -- 2.2 Simulation-Based Fault Injection -- 2.3 Fast Fault Injection for Processor Cores -- 2.3.1 Multi-Level Fault Injection -- 2.3.2 Switch from ISS Mode to Flip-Flop-Level Simulation -- 2.3.3 Switch from Flip-Flop-Level Simulation Back to ISS Mode -- 2.4 Fast Fault Injection in Uncore Components -- 2.5 Fast Fault Injection for SRAM Memories Using Mixture Importance Sampling -- 3 Cross-Layer Exploration of Soft Error Resilience Techniques -- 3.1 CLEAR: Cross-Layer Resilience for Custom Processors -- 3.1.1 Reliability Analysis -- 3.1.2 Execution Time Evaluation -- 3.1.3 Physical Design Evaluation -- 3.1.4 Resilience Library -- 3.1.5 Exploration -- 3.2 Resilience Exploration for Custom Accelerators -- 3.3 Cross-Layer Resilience for Exploration for SRAM Memories -- 3.4 Towards Cross-Layer Resiliency for Cyber-Physical Systems (CPS) -- 4 Experimental Results -- 4.1 Accuracy of FI at Different Abstraction Levels -- 4.2 Cross-Layer Resilience Exploration with CLEAR -- 4.3 Resilience Exploration for Custom Accelerators -- 4.4 Resilience Exploration for Fixed-hardware Micro-Controller -- 4.5 Resilience Exploration for SRAM Cache of Self-Balancing Robot -- 5 Conclusions -- References -- Online Test Strategies and Optimizations for Reliable Reconfigurable Architectures -- 1 Introduction and Motivation -- 1.1 Application Model -- 1.2 Runtime-Reconfigurable Architectures -- 2 Fault Detection Through Strategic Online Testing -- 2.1 Generation and Runtime Scheduling of Online Tests -- 2.2 Online Test Integration -- 2.3 Experimental Evaluation -- 3 Self-Repair by Module Diversification.

3.1 Diversified Configurations.