Chainable Compute Deployment Across Constrained Industrial Devices

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Industrial automation systems face challenges in identifying and resolving device issues due to insufficient resources for executing container-based data processing tasks, as devices may lack the necessary compute power, memory, or network bandwidth to run containers locally.

Innovation Solution

A distributed chainable compute system is deployed, where data processing tasks are distributed across available computing resources using machine learning and historical data analysis to identify suitable devices, generating a data processing flow that assigns tasks based on resource capabilities, and containers are deployed to execute these tasks, with monitoring and replacement mechanisms to ensure continuity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If containers are deployed to devices to identify and remedy conditions, then diagnostic and remediation capabilities are improved, but resource availability on constrained devices deteriorates

Engineering Contradiction:
Improvediagnostic and remediation capabilitiesVSAvoidresource availability
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent segments the container deployment strategy by device capability. Containers are deployed only to devices that have sufficient resources, while devices with constrained resources have their conditions monitored and managed by other devices in the network that can execute the containers. This segmentation resolves the contradiction by ensuring diagnostic capabilities are available without overloading constrained devices.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces intermediary devices that act as mediators between constrained devices and the container execution environment. When a constrained device experiences a condition, the system identifies an intermediary device with adequate resources to deploy and execute containers for diagnosis and remediation, allowing the constrained device to benefit from container-based solutions without directly hosting them.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If more data processing tasks are assigned to devices, then operational efficiency is improved, but device resource capacity is exceeded

Engineering Contradiction:
Improveoperational efficiencyVSAvoiddevice resource capacity
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent extends the resource dimension from individual device capacity to network-wide available capacity. By considering the collective computing resources of all devices in the network rather than relying solely on local device capacity, the system can assign and execute data processing tasks across the network, thereby improving operational efficiency without exceeding any single device's resource capacity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP4345621A1Systems and methods for condition-based deployment of chainable compute operations for constrained computing devices
Publication Date: 2024.04.03 ROCKWELL AUTOMATION TECH INC
  • EP4345621A1 patent drawingFigure 1
  • EP4345621A1 patent drawingFigure 2~3
  • EP4345621A1 patent drawingFigure 4

AI summary

A non-transitory computer readable medium stores instructions that cause a processor to receive an indication of an event associated with an industrial automation system, wherein the industrial automation system includes a plurality of devices configured to perform a plurality of operations within the industrial automation system, and wherein each of the plurality of devices includes a compute surface configured to perform one or more software tasks, determine a plurality of data processing tasks to perform based on the event, identify a portion of the plurality of devices to perform the plurality of data processing tasks based on the compute surface available for each of the plurality of devices, and deploy a container to each of the portion of the plurality of devices, wherein each container of each of the portion of the plurality of devices is configured to perform at least one of the plurality of data processing tasks.