Abstract:Internet of Things (IoT) devices are rapidly expanding in many areas, including deep mines, space, industrial environments, and health monitoring systems. Most of the sensors and actuators are battery-powered, and these batteries have a finite lifespan. Maintaining and replacing these many batteries increases the maintenance cost of IoT systems and causes massive environmental damage. Energy-harvesting devices (EHDs) are the alternative and promising solution for these battery-operated IoT devices. These EHDs collect energy from the environment and use it for daily computations, like collecting and processing data from the sensors and actuators. Using EHDs in IoT reduces overall maintenance costs and makes the IoT system energy-sufficient. However, energy availability from these EHDs is unpredictable, resulting in frequent power failures. Most of these devices use volatile memories as storage elements, implying that all collected data and decisions made by the IoT devices are lost during frequent power failures, resulting in two possible overheads. First, the IoT device must execute the application from the beginning whenever power comes back. Second, IoT devices may make wrong decisions by considering incomplete data, i.e., data-inconsistency issues. To address these two challenges, a computing model is required that backs up the collected data during power failures and restores it for later computations; this type of computing is defined as intermittent computing. However, this computing model doesn't work with conventional processors or memories. Non-volatile memory and processors are required to design a battery-less IoT device that supports intermittent computing.

Optimal Checkpointing for Secure Intermittently-Powered IoT Devices

Efficient Checkpoint under Unstable Power Supplies on NVM Based Devices.

Avoiding Data Inconsistency in Energy Harvesting Powered Embedded Systems.

DiCA: A Hardware-Software Co-Design for Differential Checkpointing in Intermittently Powered Devices

Approxify: Automating Energy-Accuracy Trade-offs in Batteryless IoT Devices

OPCIO: Optimizing Power Consumption for Embedded Devices via GPIO Configuration

Checkpoint Aware Hybrid Cache Architecture for NV Processor in Energy Harvesting Powered Systems.

iFKVS: Lightweight Key-Value Store for Flash-Based Intermittently Computing Devices

Deep-Sleep for Stateful IoT Edge Devices

Reliable and Efficient Parallel Checkpointing Framework for Nonvolatile Processor with Concurrent Peripherals

Energy versus Output Quality of Non-volatile Writes in Intermittent Computing

The Checkpoint Interval Optimization of Kernel-Level Rollback Recovery Based on the Embedded Mobile Computing System

A Dual-Threshold Scheme along with Security Reinforcement for Energy Efficient Nonvolatile Processors

Checkpointing-Aware Loop Tiling for Energy Harvesting Powered Nonvolatile Processors.

Fixing The Broken Timemachine: Consistency-Aware Checkpointing For Energyharvesting Powered Non-Volatile Processor

CP-FPGA: Computation Data-Aware Software/hardware Co-Design for Nonvolatile FPGAs Based on Checkpointing Techniques

A Novel STT-RAM-Based Hybrid Cache for Intermittently Powered Processors in IoT Devices

Design of Energy Harvesting based Hardware for IoT Applications

Optimizing Checkpoint for Scientific Simulations.

CP-FPGA: Energy-Efficient Nonvolatile FPGA with Offline/Online Checkpointing Optimization

JASS: A Flexible Checkpointing System for NVM-based Systems