Abstract:Recent embedded systems are designed with high-performance System-on-Chips (SoCs) to satisfy the computational needs of complex applications widely used in real life, such as airplane controllers, autonomous driving automobiles, medical devices, drones, and hand-held devices. Modern SoCs integrate multi-core CPUs and various types of accelerators including GPUs and DSPs. Uncontrolled heat dissipation is one of the main sources of interference that can adversely affect the reliability and real-time performance of safety-critical applications. The mechanisms currently available to protect SoCs from overheating, such as frequency throttling or core shutdown, may exacerbate the problem as they cause unpredictable delay and deadline misses. Dynamic changes in ambient temperature further increase the difficulty of solving this problem. This dissertation addresses the challenges caused by thermal interference in real-time mixed-criticality systems built with heterogeneous embedded SoC platforms. We propose a novel thermal-aware system framework with analytical timing and thermal models to guarantee safe execution of real-time tasks under the thermal constraints of a multi-core CPU/GPU integrated SoC. For mixed-criticality tasks, the proposed framework bounds the heat generation of the system at each criticality level and provides different levels of assurance against ambient temperature changes. In addition, we propose a data-driven thermal parameter estimation scheme that is directly applicable to MCSs built with commercial-off-the-shelf multi-core processors to obtain a precise thermal model without using special measurement instruments or access to proprietary information. The practicality and effectiveness of our solutions have been evaluated using real SoC platforms and our contributions will help develop systems with thermal safety and real-time predictability.

Fast Thermal Sensor Allocation Algorithms for Overheating Detection of Real Microprocessors

Temperature sensor allocation strategy and full thermal reconstruction for microprocessors

Thermal camera networks for large datacenters using real-time thermal monitoring mechanism

Power-and thermal-aware mapping for NoC under latency constraint

Research of thermal sensor allocation and placement based on dual clustering for microprocessors

Sensor Allocation Based on the Maximum Error in Thermal Monitoring for Microprocessors

A Nearest-Neighbor-Based Thermal Sensor Allocation and Temperature Reconstruction Method for 3-D NoC-Based Multicore Systems

Accurate On-Chip Thermal Peak Detection Based on Heuristic Algorithms and Embedded Temperature Sensors

Temperature Estimation Model and Applications to Thermal-Aware Test Scheduling

Reducing the Number of Sensors under Hot Spot Temperature Error Bound for Microprocessors Based on Dual Clustering

Inverse Distance Weighting Method Based on a Dynamic Voronoi Diagram for Thermal Reconstruction with Limited Sensor Data on Multiprocessors

Thermal-Aware Task Scheduling for 3D Multicore Processors.

Regulating CPU Temperature With Thermal-Aware Scheduling Using a Reduced Order Learning Thermal Model

Temperature-aware microarchitecture

Thermal-aware Resource Usage Allocation

Self-Adapting Power Density Sampling for On-Chip Thermal Prediction With Transistor-Level Granularity

Data-driven Sensor Placement for Efficient Thermal Field Reconstruction

Thermal Safety and Real-Time Predictability on Heterogeneous Embedded SoC Platforms

Improved Thermal Reconstruction Method Based on Dynamic Voronoi Diagram with Non-uniform Sampling on Multicore Processors

Peak Temperature Minimization via Task Allocation and Splitting for Heterogeneous MPSoC Real-Time Systems

Efficient Implementation of Thermal-Aware Scheduler on a Quad-core Processor