Server Power Throttling From Grid Frequency and Voltage State
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 lack precise, dynamic control over IT load adaptation.
Innovation Solution
A method where each computing server independently monitors and adapts its power consumption based on instantaneous electrical power availability from an external grid, adjusting task execution to maintain, increase, or reduce power usage, using voltage or frequency measurements to determine the power supply level and throttle resources accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If server clusters are shut down to regulate AC power frequency, then grid stability is improved, but computing server performance deteriorates
Solution Approach 1:
The patent divides the data center into multiple independent server clusters, each capable of autonomous power consumption adjustment. Instead of shutting down entire server clusters, individual servers or processing units within clusters can independently throttle their power consumption based on grid conditions, allowing finer-grained control that maintains grid stability while preserving overall computing performance.
Solution Approach 2:
The patent implements dynamic power consumption adjustment where servers can continuously adapt their power usage based on real-time grid frequency measurements. The system dynamically transitions between different operational states (full power, throttled power, standby) based on grid conditions, enabling flexible response that maintains stability without permanent performance loss.
2Loss of energy
If whole server clusters are shut down during power shortages, then power consumption is reduced, but load control precision deteriorates
Solution Approach 1:
The patent segments the computing load into independent controllable units (individual servers or processing units) rather than treating entire server clusters as single blocks. This segmentation enables precise control of power consumption at the individual unit level, allowing the system to make small incremental adjustments to match available power supply without shutting down large portions of the infrastructure.
Solution Approach 2:
The patent changes the operational parameters of computing units dynamically based on available power supply measurements. Instead of binary on/off control, the system adjusts processing speed, task execution priority, and power consumption levels as continuous variables, enabling precise matching of power demand to supply while maintaining operational flexibility.
3Stability of the object's composition
If centralized load control is implemented, then grid frequency regulation is improved, but system complexity increases
Solution Approach 1:
The patent enables each server or processing unit to autonomously monitor grid frequency and adjust its own power consumption without requiring centralized control. Each computing unit acts as an independent agent that self-regulates based on local measurements, eliminating the need for complex centralized control systems while achieving effective frequency regulation through distributed coordination.
Solution Approach 2:
The patent implements local feedback loops where each server continuously monitors grid frequency and adjusts its power consumption accordingly. The feedback mechanism is simple and distributed, with each unit responding to frequency deviations independently, reducing overall system complexity compared to centralized control while maintaining effective frequency regulation.
Data Source
AI summary
A method of adapting the power consumption of a computer, which is supplied with power from a power bus of an external electrical grid, to the power available from the power bus, wherein the computer has a processor configured for executing computing tasks, and wherein the power bus is either an AC bus or a DC bus. The method includes measuring the frequency F on the power bus in case of an AC power bus, or the voltage U on the power bus in case of a DC power bus, thus obtaining a power supply measurement indicative of the level of power supply from the power bus; and adapting the processor's computing task execution based on the obtained power supply measurement to thus maintain, increase, or reduce the computer's power consumption.