Thermal Workload Scheduling in Heterogeneous Multi-Processor SoC
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Portable computing devices (PCDs) face challenges in thermal management due to limited space, where heterogeneous processing components have varying efficiencies based on temperature, making it difficult to balance thermal energy generation and quality of service (QoS) without active cooling systems.
Innovation Solution
Implement a thermally aware workload scheduling system that monitors temperature readings of processing components, queries performance curves to compare their processing efficiencies, and allocates workloads in real-time to the most efficient components, optimizing QoS by selecting the best processing component for each task based on current and predicted thermal conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Temperature
If thermal management techniques are applied to mitigate thermal energy generation, then temperature is controlled, but quality of service deteriorates due to processor wilting or shutdown
Solution Approach 1:
The system dynamically adjusts workload allocation among processing components based on real-time temperature monitoring and performance curve analysis. The scheduler continuously adapts to changing thermal conditions, assigning tasks to the most efficient available processor rather than using a static allocation strategy, thereby maintaining productivity while managing temperature.
Solution Approach 2:
The system changes the operational parameters of processing components by selecting different processors for workload execution based on their current temperature and performance characteristics. Performance curves stored in the database allow the system to evaluate how each processor will perform under different thermal conditions and adjust workload parameters accordingly.
2Object-affected harmful factors
If spatial arrangement of components is used to manage thermal energy, then thermal interference is reduced, but device size increases due to premium on component space
Solution Approach 1:
The performance curve database acts as an intermediary between temperature monitoring and workload scheduling. Rather than physically separating components to manage thermal interference, the system uses performance curves as a mediator to intelligently allocate workloads, allowing compact component arrangement while maintaining thermal management through software-based optimization.
3Productivity
If heterogeneous processing components are used to improve processing efficiency, then productivity increases, but thermal management complexity increases due to varying efficiencies at different temperatures
Solution Approach 1:
The system performs preliminary action by pre-storing performance curves for each processing component at various temperatures in a database. This advance preparation allows the scheduler to quickly query and compare performance characteristics without complex real-time calculations, reducing the complexity of thermal management while maintaining high productivity through informed workload allocation decisions.
Data Source
AI summary
Various embodiments of methods and systems for thermally aware scheduling of workloads in a portable computing device that contains a heterogeneous, multi-processor system on a chip (“SoC”) are disclosed. Because individual processing components in a heterogeneous, multi-processor SoC may exhibit different processing efficiencies at a given temperature, and because more than one of the processing components may be capable of processing a given block of code, thermally aware workload scheduling techniques that compare performance curves of the individual processing components at their measured operating temperatures can be leveraged to optimize quality of service (“QoS”) by allocating workloads in real time, or near real time, to the processing components best positioned to efficiently process the block of code.