Data Center Computer Frequency and Cooling Adjustment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Data centers face significant challenges in managing power consumption and heat removal, leading to high operational costs due to the power-intensive nature of microprocessors and the need for extensive cooling systems, which also affects the performance and reliability of computers.

Innovation Solution

Implementing a system that monitors and adjusts the operating power and temperature of rack-mounted computers in data centers by adjusting the flow of cooling fluid and operating frequency, optimizing cooling airflow or liquid circulation to maintain optimal temperatures and reduce power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If the operating frequency of microprocessors is increased to improve computing performance, then the processing speed is improved, but the power consumption and heat generation increase

Engineering Contradiction:
Improveprocessing speedVSAvoidpower consumption
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

The system dynamically adjusts the operating frequency of microprocessors based on real-time temperature monitoring and cooling conditions. When temperatures are low, the system can operate at higher frequencies for maximum performance. When temperatures rise, the system automatically reduces frequency to prevent thermal runaway, creating a dynamic balance between performance and thermal management

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The invention changes the operating parameters (frequency) of the microprocessor based on temperature conditions. By monitoring temperature and adjusting frequency accordingly, the system optimizes the trade-off between computing performance and power consumption, allowing higher frequencies when thermally favorable and reducing frequency when heat generation becomes problematic

Inventive Principle:
Principle #35Parameter changes

2Speed

If the operating frequency is increased to improve performance, then the processing speed is improved, but the microprocessor temperature increases

Engineering Contradiction:
Improveprocessing speedVSAvoidmicroprocessor temperature
Core Design Contradiction:
SpeedVSTemperature

Solution Approach 1:

The system implements feedback control by continuously monitoring microprocessor temperature and using this information to adjust the operating frequency. Temperature sensors provide real-time data to the control system, which then modulates the frequency to maintain temperatures within safe operating limits while maximizing performance when thermally acceptable

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The operating frequency is made dynamic rather than fixed, allowing the system to adapt to changing thermal conditions. The frequency adjusts automatically in response to temperature variations, creating a dynamic equilibrium between performance delivery and thermal management

Inventive Principle:
Principle #15Dynamics

3Temperature

If cooling systems are increased to remove heat, then the temperature control is improved, but the energy consumption for cooling increases

Engineering Contradiction:
Improvetemperature controlVSAvoidcooling energy consumption
Core Design Contradiction:
TemperatureVSUse of energy by stationary object

Solution Approach 1:

Instead of providing maximum cooling capacity continuously, the system applies cooling only when and where needed. By monitoring individual microprocessor temperatures and adjusting frequency accordingly, the system avoids excessive cooling of entire racks when only specific units require attention, reducing overall cooling energy consumption

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The cooling strategy becomes localized rather than uniform. By monitoring and controlling temperature at the individual microprocessor or module level, the system can apply cooling resources precisely where heat generation is problematic, rather than cooling entire racks uniformly, thereby reducing total cooling energy requirements

Inventive Principle:
Principle #3Local quality

4Use of energy by moving object

If the operating frequency is reduced to lower power consumption, then the energy efficiency is improved, but the processing speed decreases

Engineering Contradiction:
Improvepower consumptionVSAvoidprocessing speed
Core Design Contradiction:
Use of energy by moving objectVSSpeed

Solution Approach 1:

The system dynamically adjusts frequency based on real-time thermal and workload conditions. When cooling is effective and temperatures are low, the system can operate at higher frequencies for maximum performance. When temperatures rise or cooling capacity is limited, the system automatically reduces frequency to manage thermal load, creating a dynamic optimization that balances performance and energy consumption

Inventive Principle:
Principle #15Dynamics

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach minimizes cooling costs, enhances computer performance, and ensures more uniform performance distribution across the data center by dynamically adjusting cooling resources based on the specific needs of individual computers, thereby improving overall efficiency and reducing energy expenditure.

Implementation Method 1

adjusting a flow of cooling fluid circulated to cool the computer

Methodology Applied
Scientific EffectThermal conduction: Conduction (thermal)

Implementation Method 2

adjusting a flow of cooling fluid circulated to cool the computer

Methodology Applied
Scientific EffectConvection: Convection

Data Source

PatentUS9158345B1Managing computer performance
Publication Date: 2015.10.13 GOOGLE LLC
  • US9158345B1 patent drawing
  • US9158345B1 patent drawing
  • US9158345B1 patent drawing

AI summary

Techniques for managing performance of a computing system include monitoring an operating power and an operating temperature of a rack-mounted computer in a data center; determining that the operating power is at or near a threshold operating power of the computer; and adjusting the operating frequency of the computer based on the adjusted operating temperature of the computer.