Dynamic Thread Scheduling Quantum Adjustment

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current operating systems face challenges in dynamically allocating scheduler quantum values to ensure efficient execution of threads, as fixed values do not account for varying demands of concurrently running applications, leading to potential system crashes or sluggish performance.

Innovation Solution

A dynamic thread scheduling system that determines scheduler quantum values based on real-time processing of thread performance metric data using a scheduling quantum optimizer, which adjusts values according to performance objectives and metrics, such as decode time for multimedia applications, to optimize concurrent execution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a fixed scheduler quantum value is provided by application designers, then the application can execute efficiently with a predetermined time allocation, but other concurrently running applications may not adequately perform and the system may crash due to lack of adaptability

Engineering Contradiction:

Improveapplication execution efficiencyVSAvoidscheduling adaptability to varying application demands

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamic adjustment of scheduler quantum values based on real-time system state and application performance metrics. The system continuously monitors execution patterns and modifies quantum allocations dynamically, transitioning from static fixed values to adaptive dynamic values that respond to changing conditions, thereby resolving the contradiction between efficiency and adaptability

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the scheduler quantum parameter based on observed performance metrics and system state. By monitoring execution time, resource usage, and application behavior, the system adjusts the quantum value parameter in real-time, allowing the same scheduler to optimize for different applications and conditions, thus achieving both efficiency and versatility

Inventive Principle:

Principle #35Parameter changes

2Reliability

If the scheduler quantum value is increased to allow sufficient execution time for multimedia applications, then the application can execute properly, but other concurrently running applications appear sluggish and the system may crash

Engineering Contradiction:

Improvemultimedia application performanceVSAvoidconcurrent application execution speed

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent applies different scheduler quantum values to different applications based on their specific performance requirements. Multimedia applications receiving larger quanta when needed, while other applications receive appropriate allocations, creating localized optimization for each application's quality of service rather than a uniform system-wide approach

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system provides excessive quantum allocation to multimedia applications only when performance degradation is detected, rather than continuously allocating maximum resources. This partial action approach ensures reliable multimedia playback while maintaining system responsiveness for other applications during normal operation

Inventive Principle:

Principle #16Partial or excessive action

3Productivity

If the scheduler quantum value is decreased to improve system responsiveness, then concurrent applications execute faster, but multimedia applications do not have sufficient time to execute properly

Engineering Contradiction:

Improvesystem responsivenessVSAvoidmultimedia application execution quality

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system periodically monitors multimedia application performance metrics and adjusts scheduler quantum values in response to detected degradation. This periodic feedback mechanism allows the system to maintain high responsiveness during normal operation while intervening with increased quantum allocation when multimedia execution quality deteriorates

Inventive Principle:

Principle #19Periodic action

4Device complexity

If a single fixed scheduler quantum value is used for all applications, then the scheduler implementation is simple, but the system cannot accommodate the wide variety of different application types and their varying performance requirements

Engineering Contradiction:

Improvescheduler implementation complexityVSAvoidaccommodation of diverse application requirements

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The system implements self-service scheduling where applications effectively request appropriate quantum allocation through their execution patterns and performance metrics rather than requiring manual configuration. The scheduler autonomously observes application behavior and automatically adjusts quantum values, maintaining simplicity while achieving high adaptability to diverse application types

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS7458077B2System and method for dynamically adjusting a thread scheduling quantum value

Publication Date: 2008.11.25 INTEL CORP

AI summary

A method and system optimizes thread scheduling by receiving thread objective data that includes a performance objective, sampling performance metric data points, wherein each data point varies as a function of scheduling quantum values, and calculating a new scheduling quantum value by processing the performance metric data points according to the performance objective. A process manager adjusts the scheduling quantum value to increase thread performance by utilizing the calculated scheduling quantum value.