Dynamic Power Estimation via Clock Gating Statistics

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for estimating dynamic power consumption in System on Chip (SOC) technologies suffer from low accuracy, which hampers effective temperature scheduling and heat dissipation optimization.

Innovation Solution

A dynamic power consumption estimation method and apparatus that records statistical values of target signals after clock gating, calculates dynamic power consumption based on these values, and uses weight coefficients to improve accuracy, while asynchronously counting signals to conserve resources and reduce complexity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If dynamic power consumption is obtained through static conversion using voltage, frequency, or load ratio, then the measurement process is simple, but the accuracy of dynamic power consumption estimation is low

Engineering Contradiction:
Improveaccuracy of dynamic power consumption estimationVSAvoidcomplexity of power consumption measurement system
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces traditional static electrical conversion methods with a statistical counting approach. Instead of using voltage, frequency, or load ratio conversions, the system counts the number of times clock signals are gated off to estimate dynamic power consumption. This substitution of measurement methodology significantly improves accuracy while avoiding the complexity of precise electrical measurements.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent creates a statistical model that copies the behavior of dynamic power consumption through clock gating events. By counting the frequency and duration of clock gating occurrences, the system creates a representative model of actual power consumption patterns without requiring direct measurement of power signals, thus improving accuracy without proportional increases in system complexity.

Inventive Principle:
Principle #26Copying

2Measurement precision

If clock gating signals are counted to reflect actual toggling status, then dynamic power consumption accuracy is improved, but resource consumption and system complexity increase

Engineering Contradiction:
Improveaccuracy of dynamic power consumption estimationVSAvoidresource consumption for signal counting
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent implements a self-service mechanism where the clock gating infrastructure serves dual purposes: its original function of power optimization and the additional function of power consumption measurement. The existing clock gating signals and control logic are repurposed to provide measurement data, eliminating the need for separate measurement hardware and reducing additional resource consumption.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent makes the clock gating system multi-functional by using it both for its primary purpose of reducing dynamic power consumption and for measuring power consumption statistics. This universal usage of the same hardware resources for both optimization and measurement functions minimizes additional resource requirements while achieving high measurement accuracy.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS12099392B2Dynamic power consumption estimation method, apparatus, and system
Publication Date: 2024.09.24 HUAWEI TECH CO LTD
  • US12099392B2 patent drawing
  • US12099392B2 patent drawing

AI summary

A dynamic power consumption estimation method includes recording a statistical value of each of n target signals of a to-be-measured target in a unit time, where the target signal is a clock signal obtained after clock gating and the statistical value is a quantity of times toggling, or the target signal is a clock gating signal and the statistical value is a quantity of enable cycles, and calculating the dynamic power consumption of the to-be-measured target in the unit time based on the statistical value of each of the n target signals in the unit time.