Performance-Hint Resource Allocation for Gaming Frame Stability

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing resource scaling and allocation schemes rely on manual workload evaluation and user/sensor inputs, leading to an inability to achieve a good power-performance balance, often resulting in performance sacrifices such as frame drops in gaming applications.

Innovation Solution

A performance-hint-driven dynamic resource management system that receives workload requirements and sensor inputs, determines new resource allocations, reconfigures resources, evaluates performance, and generates performance hints to optimize power and performance balance by adjusting processor frequency and workload distribution between processors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Use of energy by stationary object

If resources are scaled based on past workload using manual evaluation, then power consumption is reduced, but performance stability deteriorates causing frame drops

Engineering Contradiction:

Improvepower consumptionVSAvoidperformance stability

Core Design Contradiction:

Use of energy by stationary objectVSReliability

Solution Approach 1:

The system implements a feedback mechanism where performance metrics (frame rates, rendering times) are continuously monitored and fed back to the resource allocation unit. This closed-loop feedback enables dynamic adjustment of resource allocation to maintain performance stability while optimizing power consumption, resolving the contradiction between power reduction and performance reliability.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent transitions from static manual resource evaluation to dynamic automated resource allocation. The system continuously monitors workload patterns and performance metrics, dynamically adjusting resource allocation in real-time. This dynamic approach prevents frame drops by adapting to changing conditions while maintaining power efficiency, resolving the contradiction between power consumption and performance stability.

Inventive Principle:

Principle #15Dynamics

2Device complexity

If resources are allocated manually based on periodic evaluation, then device complexity is reduced, but adaptability to workload changes deteriorates

Engineering Contradiction:

Improveallocation mechanism complexityVSAvoidworkload adaptability

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The system implements self-service automation where the resource allocation unit autonomously monitors workload patterns, evaluates performance metrics, and adjusts resource allocation without manual intervention. This self-service mechanism enhances adaptability to workload changes while managing complexity through automated decision-making algorithms, resolving the contradiction between device complexity and workload adaptability.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system performs preliminary actions by pre-configuring resource allocation strategies based on predicted workload patterns and performance requirements. The automated evaluation unit proactively adjusts resources before performance degradation occurs, enabling the system to adapt to workload changes without complex real-time manual adjustments, thus resolving the contradiction between complexity and adaptability.

Inventive Principle:

Principle #10Preliminary action

3Reliability

If processor frequency is increased to maintain frame rate, then performance stability is improved, but power consumption increases

Engineering Contradiction:

Improveframe rate stabilityVSAvoidpower consumption

Core Design Contradiction:

ReliabilityVSUse of energy by stationary object

Solution Approach 1:

The system dynamically changes operational parameters (processor frequency, workload distribution) based on real-time performance feedback and workload analysis. By intelligently adjusting these parameters rather than maintaining fixed high frequency, the system maintains frame rate stability while minimizing power consumption, resolving the contradiction between frame rate stability and power usage.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent segments the processing workload across multiple processors or cores, allowing the system to maintain performance stability through distributed processing rather than relying on a single high-frequency processor. This segmentation enables better power management by activating only the necessary processing units, resolving the contradiction between frame rate stability and power consumption.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS10514956B2Performance hint driven resource allocation

Publication Date: 2019.12.24 QUALCOMM INC

AI summary

Performance-hint-driven dynamic resource management, including: receiving workload requirements and sensor inputs of a system; determining a new allocation for resources of the system; reconfiguring the resources of the system using the new allocation; evaluating performance of the system based on the reconfigured resources of the system; and generating performance hints based on the evaluated performance of the system.