Data Center Power Analysis System for Energy Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current power analysis technologies for data centers lack a holistic approach to achieve energy efficiency while meeting business needs, resulting in inefficient energy consumption.

Innovation Solution

A method and system for power analysis that involves receiving and processing information from various data center components, estimating power usage based on stored data and measurements, and providing recommendations for improving energy efficiency by identifying unutilized components, consolidating resources, and optimizing thermal profiles using Computational Fluid Dynamics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If data centers increase computing capacity and components to meet growing business demand, then productivity and service capability improve, but energy consumption increases significantly

Engineering Contradiction:
Improvecomputing capacityVSAvoidenergy consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The system changes operational parameters by dynamically adjusting power states of components based on actual utilization metrics. It monitors CPU usage, memory access patterns, storage I/O rates, and network traffic to determine optimal power states, transitioning components between active, idle, and sleep states to reduce energy consumption while maintaining required computing capacity.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system identifies and power-cycles components that are in unutilized or low-utilization states. By detecting components that have been idle beyond a threshold period, the system safely powers them down to save energy, and automatically powers them back up when utilization is detected, thus recovering energy without impacting active workloads.

Inventive Principle:
Principle #34Discarding and recovering

2Loss of energy

If data centers implement comprehensive power monitoring and analysis systems, then energy efficiency improves, but device complexity increases

Engineering Contradiction:
Improveenergy efficiencyVSAvoidsystem complexity
Core Design Contradiction:
Loss of energyVSDevice complexity

Solution Approach 1:

The system enables components to self-report their utilization status through embedded sensors and monitoring agents that collect metrics on CPU usage, memory access, storage I/O, and network traffic. Components autonomously determine when they are unutilized and signal this state to the power management system, eliminating the need for complex centralized monitoring of every component state.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system implements continuous feedback loops where power state changes are monitored and correlated with utilization metrics. This feedback mechanism validates whether power cycling decisions were appropriate and adjusts future decisions accordingly, ensuring energy efficiency improvements while maintaining system stability and preventing excessive complexity.

Inventive Principle:
Principle #23Feedback

3Loss of energy

If data centers power cycle components to reduce energy consumption, then energy efficiency improves, but reliability may be affected

Engineering Contradiction:
Improveenergy consumptionVSAvoidsystem reliability
Core Design Contradiction:
Loss of energyVSReliability

Solution Approach 1:

The system performs preliminary actions by gradually transitioning components through intermediate power states (e.g., from active to idle to sleep) rather than abrupt on/off switching. It pre-warms components before full activation and implements gradual power ramping to minimize thermal shock and electrical stress, thereby maintaining reliability while achieving energy savings.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements cushioning measures by monitoring component health status and operational patterns before power cycling. It identifies components with stable, predictable workloads that are safe to power cycle, while protecting components with critical or volatile workloads. The system also implements retry logic and health checks after power cycling to ensure reliable recovery.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

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 enables data centers to optimize power usage and thermal management, leading to reduced energy consumption and improved efficiency by identifying areas for improvement and implementing strategic changes such as eliminating unutilized components and optimizing component placement.

Implementation Method 1

estimating a thermal profile of a component based at least in part on one or more of the estimated power usage of the component and one or more received measurements

Methodology Applied
Scientific EffectThermal conduction: Conduction (thermal)

Data Source

PatentEP2457153B1Method and system for power analysis
Publication Date: 2016.11.02 SCHNEIDER ELECTRIC IT CORP
  • EP2457153B1 patent drawingFigure 1
  • EP2457153B1 patent drawingFigure 2
  • EP2457153B1 patent drawingFigure 3

AI summary

Techniques for power analysis for data centers are disclosed In one particular exemplary embodiment, the techniques may be realized as a method for power analysis for a plurality of computing platform components comprising receiving information associated with a component, retrieving, using a computer processor, electronically stored data associated with the component, estimating power usage of the component based at least in part on the stored data, and outputting an indicator of power usage.