Proxy-Based Instruction Throttling for Multicore Voltage Droop
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern multicore processors face energy management challenges due to voltage droop and heat generation issues, leading to potential microarchitectural stall events and reduced lifespan, as they operate with multiple cores powered by a steady-state power source, which can result in significant voltage drops and thermal excursions.
Innovation Solution
A system and method utilizing a power state general purpose engine and hardware-based control loop to manage energy consumption by establishing temporal intervals and adjusting processor core throttling states, with a digital power proxy approximating real-time energy consumption and a proportional-integral control loop for fine-grained power management, ensuring convergence of total period energy consumption with interval targets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple processor cores are powered through a steady-state power source, then processing capacity is increased, but voltage droop occurs due to increased loading
Solution Approach 1:
The patent implements dynamic power management by dividing the temporal interval into multiple temporal periods and adjusting the processor core throttling state for each period based on actual energy consumption. This dynamic adjustment allows the system to adapt to varying processing loads and prevent voltage droop while maintaining high processing capacity.
Solution Approach 2:
The patent employs a feedback mechanism where the actual energy consumption of processor cores is monitored and used to adjust the throttling state for subsequent temporal periods. This closed-loop control ensures that power consumption remains within safe limits, preventing voltage droop and microarchitectural stall events.
2Productivity
If processor cores operate at high performance, then productivity is improved, but heat generation increases causing thermal excursions
Solution Approach 1:
The patent applies periodic action by dividing operation into temporal intervals and temporal periods, with throttling states adjusted at the beginning of each successive temporal period. This periodic control allows the processor to operate at high performance during cooler periods while preventing thermal excursions through controlled throttling when temperature thresholds are approached.
3Use of energy by moving object
If processor core throttling is increased to reduce power consumption, then energy management is improved, but processing activity is reduced
Solution Approach 1:
The patent implements dynamic throttling adjustment where the throttling state is modified for each temporal period based on actual energy consumption patterns. This allows the system to minimize throttling (maximize processing activity) when energy consumption is within targets, and apply throttling only when necessary to meet energy targets, thus optimizing the balance between energy efficiency and processing productivity.
4Measurement precision
If fine-grained power management is implemented, then energy control precision is improved, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the temporal interval into multiple temporal periods and establishing separate throttling states for each period. This segmentation enables fine-grained energy control with precise monitoring and adjustment at each period boundary, achieving high measurement precision for energy consumption while managing complexity through structured temporal division.
Data Source
AI summary
Embodiments relate to a system and method for managing energy consumption of one or more processor cores in a multicore processing device. The method includes establishing a temporal interval that includes a plurality of temporal periods and an interval energy target for one or more processor cores. The method also includes determining for each temporal period a period energy target for the processor cores and determining a processor core throttling state for the processor cores. The method further includes adjusting the respective period energy target and the respective processor core throttling state at the beginning of each successive temporal period. The method also includes converging, subject to the adjusting, as each respective temporal period of the plurality of temporal periods is concluded, a total period energy consumption of the processor cores with the interval energy target.