Proactive Controller for Hard Real-Time Systems

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Managing hard real-time electronic systems with adaptive 'knobs' is challenging due to uncertainties in system dynamics, such as computation and bandwidth requirements, which affect optimality and feasibility, and requires efficient control to minimize costs and meet constraints like deadlines and quality-of-service.

Innovation Solution

A method that exploits task information determined by a predictor to select and adjust working modes based on upper bound information and constraints, using a dynamical procedure to optimize resource utilization, with a proactive controller that accounts for slack and switching overheads, and employs look-up tables for energy/execution-time trade-offs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Use of energy by moving object

If dynamic knob tuning is used to optimize system metrics, then energy consumption and quality-of-service are improved, but system complexity and control difficulty increase

Engineering Contradiction:

Improveenergy consumptionVSAvoidsystem complexity

Core Design Contradiction:

Use of energy by moving objectVSDevice complexity

Solution Approach 1:

The system performs self-adjustment through automatic knob tuning based on runtime monitoring of system metrics and uncertainty bounds, eliminating the need for manual intervention while optimizing energy consumption and quality-of-service

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system dynamically changes operational parameters (knob settings) based on runtime conditions and uncertainty analysis, allowing optimization of energy consumption and performance without requiring complex manual control

Inventive Principle:

Principle #35Parameter changes

2Reliability

If conservative design-time upper bounds are used, then constraint satisfaction is guaranteed, but resource utilization efficiency deteriorates

Engineering Contradiction:

Improveconstraint satisfactionVSAvoidresource utilization

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system pre-computes refined upper bounds at design time using uncertainty analysis, and then uses these pre-computed bounds at runtime to make efficient scheduling decisions that both satisfy constraints and improve resource utilization

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses runtime monitoring of actual system behavior against predicted uncertainty bounds to dynamically adjust scheduling decisions, ensuring constraint satisfaction while improving resource utilization through feedback-driven optimization

Inventive Principle:

Principle #23Feedback

3Adaptability or versatility

If frequent knob switching is performed, then system adaptability and optimization are improved, but switching overhead increases

Engineering Contradiction:

Improvesystem adaptabilityVSAvoidswitching overhead

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system dynamically determines when knob switching is necessary by monitoring uncertainty bounds and system metrics, adjusting the frequency of switching based on actual runtime conditions rather than using a fixed schedule

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system performs switching only when uncertainty analysis indicates it is necessary to maintain constraint satisfaction or optimize performance, avoiding unnecessary switching actions that would consume time and resources

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS8856791B2Method and system for operating in hard real time

Publication Date: 2014.10.07 INTERUNIVERSITAIR MICRO ELECTRONICS CENT (IMEC VZW)

AI summary

A method for managing the operation of an electronic system by taking into account various cost constraints of the system is disclosed. In one aspect, the method includes selecting a working mode for a plurality of tasks in a pro-active way using predictive control mechanism while guaranteeing hard real time constraints. The system is operated at the selected working mode for the corresponding tasks.