Container Resource Allocation for Low-Workload Automated Driving

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional containerized virtualization technologies face inefficiencies in utilizing high-performance hardware resources when automated driving applications experience low workload conditions, leading to idle times and underutilization.

Innovation Solution

A control device manages multiple containers to share hardware resources and a host operating system, allocating the remainder of the hardware resources to user applications when the automated driving application's workload decreases to a set range, ensuring efficient utilization of high-performance hardware.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If high-performance hardware resources are allocated to automated driving applications, then the reliability and safety of automated driving processes are improved, but the productivity and resource utilization efficiency deteriorate when workload is low

Engineering Contradiction:

Improvesafety of automated driving processVSAvoidresource utilization efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent implements dynamic resource allocation where the container management layer continuously monitors workload conditions and adjusts hardware resource allocation in real-time. When automated driving workload is high, dedicated resources ensure safety; when workload is low, resources are dynamically reallocated to user applications, resolving the contradiction between reliability and productivity

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent makes hardware resources universal by allowing them to serve multiple purposes: dedicated automated driving functions when needed, and shared user applications when automated driving workload is low. The container management layer enables this multi-functionality through intelligent resource orchestration, improving both safety and resource utilization efficiency

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Stability of the object's composition

If dedicated hardware resources are reserved for automated driving applications, then the stability and performance of automated driving processes are improved, but the adaptability and flexibility of the system deteriorate

Engineering Contradiction:

Improvestability of automated driving processVSAvoidsystem flexibility

Core Design Contradiction:

Stability of the object's compositionVSAdaptability or versatility

Solution Approach 1:

The patent segments hardware resources into dedicated portions for automated driving and shareable portions for user applications. The container management layer manages this segmentation dynamically, ensuring stable dedicated resources when needed while maintaining system flexibility through configurable resource allocation policies

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system transitions from static dedicated resource allocation to dynamic allocation based on workload conditions. The container management layer enables this dynamic behavior, allowing the system to adapt between stability-oriented and flexibility-oriented modes, resolving the contradiction between stability and adaptability

Inventive Principle:

Principle #15Dynamics

3Power

If high-performance hardware resources are allocated to automated driving applications, then the power and computational capability are improved, but the loss of energy and resource waste increase during idle times

Engineering Contradiction:

Improvecomputational capabilityVSAvoidresource waste during idle times

Core Design Contradiction:

PowerVSLoss of energy

Solution Approach 1:

The patent ensures continuous useful action by reallocating hardware resources from automated driving to user applications during low workload periods. The container management layer monitors workload continuously and maintains productive utilization of resources, eliminating idle times and reducing energy waste while preserving computational capability when needed

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

The patent implements resource recovery by temporarily releasing dedicated hardware resources from automated driving applications when workload is low and reallocating them to user applications. The container management layer manages this recovery process, reducing resource waste while maintaining the ability to recover and reallocate resources when automated driving demand increases

Inventive Principle:

Principle #34Discarding and recovering

Data Source

PatentUS20250284548A1Control device, resource management method and storage medium thereof

Publication Date: 2025.09.11 DENSO CORP

AI summary

A control device controls multiple containers to share a hardware resource and a host operating system, and includes: an automated driving container executing an automated driving application on the host operating system, the automated driving application executing an automated driving process of a vehicle; a user container executing a user application on the host operating system, the user application being designated by a user; and a container management layer managing an allocation of the hardware resource to the automated driving application and the user application. In response to determining that a low workload condition, in which a workload of the automated driving application decreases to a set range, being satisfied, the container management layer allocates, to the user application, a remainder of the hardware resource, which is required for continuing an execution of the automated driving application.