Processor power consumption optimization method suitable for Android games

Abstract

The invention discloses a processor power consumption optimization method suitable for Android games. A power consumption management framework for Android platform game applications is constructed andincludes a game characteristic data collection module and a CPU-GPU coordinated voltage frequency control module. The characteristic data collection module completes the collection of user interaction data and GPU rendering data during Android game running, provides support for CPU and GPU load evaluation based on Android game characteristics, and improves timeliness and accuracy; and the coordinated voltage frequency control module uses a hierarchical state machine to comprehensively describe game running state changes and processor running state changes and indicates the timing and direction for voltage frequency control logic. The control logic achieves CPU-GPU coordinated power consumption management based on a producer-consumer collaborative relationship between a CPU and a GPU in anAndroid game. The CPU-GPU energy efficiency is improved on the premise of ensuring user experience.

Description

technical field [0001] The invention relates to a method for optimizing power consumption of a processor, in particular to a method for optimizing power consumption of a processor suitable for Android games. Background technique [0002] With the popularity of smart phones with good human-computer interaction characteristics, mobile games have gradually become a type of application that is widely loved by users, and complex 3D games are especially popular. This is due to the improvement of the computing power of the smartphone CPU (Central Processing Unit) and GPU (Graphics Processing Unit). High performance often brings high energy consumption. For battery-powered smartphone platforms, game applications have high requirements for computing resources, which highlights the contradiction between the inherent power consumption and battery capacity of smartphone devices. The study found that there is a waste of computing resources and energy consumption when running game applic...

AI Technical Summary

Problems solved by technology

[0004] The problems that the traditional DVFS strategy is suitable for Android game applications are: (1) Android games have significant user interaction characteristics and CPU-GPU cooperation characteristics when running, and traditional frameworks are difficult to use these characteristics to optimize power consumption

(2) The CPU / GPU load evaluation method based on hardware utilization feedback cannot meet the timeliness and accuracy requirements of Android game load evaluation, resulting in reduced user experience or waste of energy consumption

(3) The completely independent CPU / GPU voltage and frequency control logic cannot use the cooperative relationship between CPU-GPU when Android games are running to further improve system energy efficiency

Images