Multi-Core Processor Core Switching and Frequency Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-core processors, high-performance cores exhibit high power consumption and heat generation, leading to instability as the number of used high-performance cores increases, which existing power management technologies like DVFS, Intel's EIST, and Turbo Boost fail to effectively manage.
Innovation Solution
A method is introduced to dynamically switch between low-power and high-performance cores based on CPU load and the number of operating high-performance cores, involving measuring CPU load, counting high-performance cores, determining maximum operating frequency, and switching contexts to optimize power usage and thermal management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If high-performance cores are used to increase processing speed, then productivity is improved, but power consumption and heat generation increase
Solution Approach 1:
The system dynamically switches between low-power cores and high-performance cores based on CPU load conditions. When CPU load is low, low-power cores are used to minimize energy consumption. When CPU load is high, high-performance cores are activated to meet processing demands, creating a dynamic adaptation to workload requirements
Solution Approach 2:
The system changes operational parameters by adjusting the number of active high-performance cores based on the counted number of currently operating high-performance cores. This parameter change allows the system to optimize the balance between processing speed and power consumption according to actual workload needs
2Productivity
If the number of high-performance cores increases to handle higher CPU load, then productivity is improved, but heat generation increases leading to instability
Solution Approach 1:
The system implements feedback control by counting the number of currently operating high-performance cores and using this information to determine the maximum operating frequency. This feedback mechanism prevents excessive heat generation by adjusting frequency based on the actual number of active high-performance cores, thereby maintaining system stability
Solution Approach 2:
The system dynamically adjusts the operating frequency of high-performance cores based on the number of active cores. This dynamic frequency adjustment ensures that the system can handle varying CPU loads while maintaining thermal stability and preventing overheating-induced instability
3Productivity
If operating frequency is increased above rated frequency to meet high CPU demand, then productivity is improved, but power consumption and heat generation increase beyond TDP limit
Solution Approach 1:
The system changes the maximum operating frequency parameter of high-performance cores based on the counted number of active cores. By dynamically adjusting this parameter, the system can achieve higher throughput when needed while ensuring that the total power consumption remains within the thermal design power limit
Solution Approach 2:
The system implements dynamic frequency scaling where the operating frequency of high-performance cores is adjusted in real-time based on the number of active cores and CPU load conditions. This dynamic adjustment allows the system to optimize performance while adhering to power and thermal constraints
Data Source
AI summary
A method of operating a system on chip (SoC) includes determining to switch from a selected low-power core among a plurality of low-power cores to a high-performance core among a plurality of high-performance cores, counting the number of high-performance cores that are operating among a plurality of high-performance cores, determining a maximum operating frequency of the plurality of high-performance cores based on the counted number, and switching from the selected low-power core to the selected high-performance core based on the determined maximum operating frequency.