Dynamic Power Limit Adjustment for Energy Efficient Computer Process

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for reducing energy consumption in computer systems are inefficient, requiring significant time and effort to classify and modify software applications, and are difficult to implement during runtime, while modern computer architectures need to balance energy usage with performance.

Innovation Solution

A process that samples energy and power measurements at regular intervals, computes metrics, classifies software samples as memory-bound or compute-bound, and adjusts hardware power limits dynamically to optimize energy utilization without compromising performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Use of energy by moving object

If software applications are classified and modified to be more energy efficient, then energy consumption is reduced, but significant time and effort are required and modifications are difficult to implement during runtime

Engineering Contradiction:

Improveenergy consumptionVSAvoidtime and effort required

Core Design Contradiction:

Use of energy by moving objectVSLoss of time

Solution Approach 1:

The system enables software applications to automatically classify themselves and have their power consumption dynamically adjusted without requiring external modification or manual intervention. The self-service mechanism allows the computer system to autonomously analyze software behavior patterns and apply appropriate power limits, eliminating the time-consuming manual classification and modification process while achieving energy efficiency goals.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent implements dynamic power limit adjustment that can be applied during runtime based on real-time software classification. Instead of static pre-modification approaches, the system continuously monitors software behavior and dynamically adjusts power consumption limits, enabling energy efficiency improvements without requiring application restarts or manual modifications, thus resolving the contradiction between energy reduction and time investment.

Inventive Principle:

Principle #15Dynamics

2Use of energy by moving object

If power consumption is reduced through software modification, then energy efficiency improves, but system performance may be compromised

Engineering Contradiction:

Improveenergy efficiencyVSAvoidsystem performance

Core Design Contradiction:

Use of energy by moving objectVSProductivity

Solution Approach 1:

The system changes power consumption parameters dynamically based on software classification results. By adjusting power limits according to the specific characteristics of each software application (e.g., memory-bound vs. compute-bound), the system optimizes energy efficiency while maintaining adequate performance levels. This parameter adjustment approach allows flexible balancing between energy consumption and performance requirements without permanent software modifications.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent implements a feedback mechanism where power consumption measurements are continuously monitored, software is classified based on its behavior patterns, and power limits are adjusted accordingly. This closed-loop feedback system ensures that energy efficiency improvements are achieved while maintaining system performance within acceptable ranges, as the feedback allows continuous optimization rather than one-time modification that might compromise performance.

Inventive Principle:

Principle #23Feedback

3Measurement precision

If energy counters and power measurements are sampled continuously, then power profiling accuracy improves, but energy consumption increases

Engineering Contradiction:

Improvepower profiling accuracyVSAvoidenergy consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system applies partial sampling action by measuring power consumption at strategically selected intervals and for specific hardware components rather than continuously monitoring everything. This partial measurement approach achieves sufficient power profiling accuracy to classify software applications correctly while avoiding the excessive energy consumption that would result from continuous comprehensive monitoring of all system components.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS10817041B2Energy efficient computer process

Publication Date: 2020.10.27 VACCARO PHILIP

AI summary

Described is a process for optimizing energy utilization in a computer processing device, including sampling energy and power measurements for given hardware components, storing the energy and power measurements, computing metrics for a sample based on current power consumption and current power limits, comparing the current sample metrics against a metric threshold, classifying the current sample based on the comparison, assigning a classification type based on the classifying the current sample, determining if an actual number of a computational intensity characteristic exceeds a maximum allowed reference number in a sample window for the classification type, if the actual number of a computational intensity characteristic exceeds the maximum for the classification type, computing a new maximum allowed for the classification type, and constraining the hardware components to the new maximum energy and power measurements.