Server Power Adaptation Using Grid Frequency and Bus Voltage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data center power management schemes, such as those that regulate AC power frequency by shutting down server clusters during power shortages, negatively impact computing server performance and are not adaptable enough to varying power supply levels.
Innovation Solution
A method where each computing server independently monitors and adapts its power consumption based on instantaneous power availability from an external electrical grid, adjusting task execution to match available power, thereby maintaining, increasing, or reducing electrical power consumption dynamically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If server clusters are shut down to regulate AC power frequency during power shortages, then grid stability is maintained, but server performance and productivity are significantly reduced
Solution Approach 1:
The patent implements dynamic power consumption adaptation where computing systems continuously monitor grid state indicators (frequency, voltage) and dynamically adjust their power consumption levels. Instead of static shutdowns, the system transitions smoothly between different operational states based on real-time grid conditions, allowing servers to operate at reduced but non-zero capacity during power shortages, thus maintaining both grid stability and continuous productivity.
Solution Approach 2:
The system changes operational parameters (power consumption level, task execution rate) based on measured grid state parameters (frequency deviation, voltage level). When grid frequency drops below threshold or voltage deviates from nominal, the computing system adjusts its power consumption parameter accordingly, creating a continuous adaptation mechanism that resolves the contradiction between grid stability requirements and server performance maintenance.
2Quantity of substance
If whole server clusters are shut down during power shortages, then power consumption is reduced to match supply, but the adaptation precision to varying power levels is lost
Solution Approach 1:
Instead of complete shutdowns (excessive action), the patent implements partial power consumption reduction where servers operate at scaled-back capacity. The system applies partial throttling of computing tasks based on grid conditions, consuming just enough power to match available supply while maintaining non-zero operational capacity, thus achieving precise adaptation across a continuous range of power availability levels.
Solution Approach 2:
The system continuously measures grid state indicators and uses this feedback to adjust power consumption levels. The monitoring of frequency and voltage provides real-time feedback about grid conditions, which triggers proportional adjustments in computing load, creating a closed-loop control system that achieves precise adaptation to varying power supply levels rather than binary on/off states.
3Stability of the object's composition
If centralized load control is implemented to manage data centre power, then grid frequency can be regulated, but system complexity and control overhead increase
Solution Approach 1:
The patent divides the load control function into independent segments at each computing system rather than implementing centralized control. Each computing system independently monitors grid state and autonomously adjusts its own power consumption, eliminating the need for complex centralized control infrastructure while achieving collective grid frequency regulation through distributed coordination.
Solution Approach 2:
Each computing system performs self-service load management by autonomously monitoring grid state indicators and independently adjusting its own power consumption without external control commands. The systems self-regulate based on predefined thresholds and algorithms, reducing control system complexity while maintaining frequency regulation capability through distributed autonomous decision-making.
Data Source
Figure 1
Figure 2
Figure 3a~3b
AI summary
A method of adapting the power consumption of a computer (102), which is supplied with power from a power bus (104) of an external electrical grid, to the power available from the power bus (104), wherein the computer (102) has a processor (106) configured for executing computing tasks, and wherein the power bus (104) is either an AC bus or a DC bus, the method comprising the following steps: a. measuring the frequency F on the power bus (104) in case of an AC power bus, or the voltage U on the power bus (104) in case of a DC power bus, thus obtaining a power supply measurement (Ubus) indicative of the level of power supply from the power bus (104); and b. adapting the processor's computing task execution based on the obtained power supply measurement (Ubus) to thus maintain, increase, or reduce the computer's power consumption.