Dynamic Power Budget Allocation Across Processor Domains
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multicore processors with varying circuitry units face challenges in ensuring each unit receives sufficient power based on its workload, as existing mechanisms are inadequate for dynamic power allocation between different domains.
Innovation Solution
A power budget management algorithm dynamically allocates power between domains by calculating a package power budget and adjusting frequency and voltage based on guaranteed and turbo frequencies, using configuration registers to set policy values that determine how the power is shared between core and graphics domains, allowing for reallocation of power headroom.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If dynamic power allocation mechanisms are implemented, then power distribution efficiency is improved, but device complexity increases
Solution Approach 1:
The processor is divided into multiple independent power domains (core domain and graphics domain), each with its own power management capabilities. This segmentation allows independent control of power allocation to different functional units, improving power distribution efficiency while managing complexity through modular domain structures.
Solution Approach 2:
The power allocation mechanism dynamically adjusts power distribution between domains based on real-time workload demands. The system continuously monitors power consumption and workload characteristics, then reallocates power budgets dynamically rather than using static allocation, thereby improving efficiency without requiring overly complex predetermined configuration schemes.
2Productivity
If power is allocated based on workload demands, then productivity is improved, but power management complexity increases
Solution Approach 1:
The power management system incorporates feedback mechanisms that monitor workload demands, power consumption levels, and performance requirements. Based on this feedback, the system automatically adjusts power allocation between core and graphics domains, ensuring high productivity while managing complexity through closed-loop control rather than open-loop complex scheduling.
Solution Approach 2:
The system changes operational parameters (frequency, voltage, power budget) dynamically based on workload conditions. By adjusting these parameters in response to measured performance needs, the system achieves high productivity without requiring complex structural changes, instead using parameter modulation to adapt to varying demands.
Data Source
AI summary
In one embodiment, the present invention includes a method for determining a power budget for a multi-domain processor for a current time interval, determining a portion of the power budget to be allocated to first and second domains of the processor, and controlling a frequency of the domains based on the allocated portions. Such determinations and allocations can be dynamically performed during runtime of the processor. Other embodiments are described and claimed.