Server Rack Power Management via Dynamic Frequency Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional server rack systems face issues where servers crash or shut down due to power overload when one or more power supply units malfunction, leading to potential information loss or damage.

Innovation Solution

A server rack system with a microcontroller that monitors the number of operational servers and power supply units, controlling servers to operate at a lower frequency when the number of operational servers exceeds a critical value and at least one power supply unit operates abnormally, thereby preventing power overload.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If servers operate at full power consumption, then productivity is improved, but reliability deteriorates when power supply units malfunction

Engineering Contradiction:
Improveserver operating performanceVSAvoidserver continuous operation
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The microcontroller continuously monitors the operational status of power supply units and adjusts server operating frequencies based on real-time power supply capacity. When a power supply unit malfunctions, the system receives feedback about the reduced power capacity and automatically adjusts server operations to match the available power, preventing crashes and ensuring continuous operation.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system dynamically adjusts server operating frequencies based on the actual power supply capacity. Instead of fixed full-power operation, servers operate at variable frequencies that adapt to the current power supply unit status. This dynamic adjustment allows the system to maintain optimal performance when full power is available while preventing overload when power supply units malfunction.

Inventive Principle:
Principle #15Dynamics

2Productivity

If the number of operational servers increases, then productivity is improved, but power consumption increases, causing servers to crash when exceeding power supply capacity

Engineering Contradiction:
Improvenumber of operational serversVSAvoidserver stability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system changes the operating frequency parameter of servers based on the number of operational servers and power supply unit status. When more servers are operational or when power supply units malfunction, the microcontroller adjusts the operating frequencies to ensure total power consumption remains within the available power supply capacity, preventing crashes while maximizing the number of operational servers.

Inventive Principle:
Principle #35Parameter changes

3Power

If power supply units operate at full capacity, then power output is improved, but servers crash when power consumption exceeds rated output power

Engineering Contradiction:
Improvepower supply outputVSAvoidserver operation continuity
Core Design Contradiction:
PowerVSReliability

Solution Approach 1:

The microcontroller monitors the rated output power of power supply units and uses this feedback to control server operating frequencies. When power supply units operate at full capacity, the system ensures that server power consumption does not exceed the rated output, preventing crashes while maximizing power utilization.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS11224141B2Server rack system and method of power management in the server rack system
Publication Date: 2022.01.11 MITAC COMPUTING TECH
  • US11224141B2 patent drawing
  • US11224141B2 patent drawing
  • US11224141B2 patent drawing

AI summary

A server rack system includes a plurality of servers, a plurality of power supply units for supplying power to the servers, and a microcontroller electrically connected to the servers and the power supply units. The microcontroller is configured to determine a number of those of the servers that are currently in operation, compare the number with a critical value, determine whether each of the power supply units operates normally, and control at least one of the servers that are currently in operation to operate at a lower operating frequency when the number is greater than the critical value and at least one of the power supply units operates abnormally.