Industrial Software Sidecar Tracking in Complex Intranet Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing industrial software tracking solutions are intrusive, require significant code modifications, are difficult to implement in complex industrial networks, and lack support for non-public network environments, leading to performance burdens and limited event tracking capabilities.
Innovation Solution
A codeless tracking analytics mechanism using a sidecar agent for industrial software, which identifies intranet topology, injects tracking points into edge devices, and performs data synchronization and analysis without modifying existing source code, utilizing a hybrid control plane and sidecar agents for non-intrusive tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If intrusive code injection methods are used for tracking, then tracking functionality can be implemented, but system complexity increases and original code must be modified
Solution Approach 1:
The patent introduces an AOP (Aspect-Oriented Programming) framework as an intermediary layer between the tracking requirements and the original software code. This framework allows tracking functionality to be injected without directly modifying the original code, thereby reducing system complexity while maintaining adaptability. The AOP framework acts as a mediator that handles the complexity of code injection centrally.
Solution Approach 2:
The patent segments the tracking system into modular components including aspect definitions, pointcuts, and advice functions. This segmentation allows the tracking functionality to be independently configured and deployed without affecting the core software architecture, thus reducing overall system complexity while maintaining versatility.
2Loss of information
If cloud-based tracking services are used, then comprehensive analytics can be achieved, but network security requirements and data privacy concerns increase
Solution Approach 1:
The patent enables tracking data to be processed and analyzed locally on the edge device rather than requiring constant cloud connectivity. This local processing capability maintains comprehensive analytics functionality while reducing network security risks and data privacy concerns by keeping sensitive information within the local environment.
Solution Approach 2:
The patent introduces a hybrid architecture that operates across two dimensions: local edge computing for real-time processing and cloud-based services for comprehensive analytics when available. This dimensional approach allows the system to achieve comprehensive analytics capability while adapting to network security requirements by switching between local and cloud-based operations.
3Loss of information
If heavy tracking solutions are implemented, then detailed user behavior analysis is possible, but system performance is burdened
Solution Approach 1:
The patent implements selective tracking by defining specific pointcuts that target only the necessary functions and events for analysis. Rather than instrumenting all code paths, the system applies tracking only where needed based on the aspect configuration, thereby maintaining detailed behavior analysis capability while minimizing the performance burden on the system.
4Ease of manufacture
If codeless tracking is implemented in industrial environments, then deployment simplicity improves, but compatibility with diverse device architectures decreases
Solution Approach 1:
The patent creates a universal AOP-based tracking framework that can operate across diverse industrial device architectures including PLCs, PACs, and industrial PCs. The framework provides multi-functional support for different programming languages and device types through a standardized interface, thereby maintaining deployment simplicity while achieving broad device architecture compatibility.
Data Source
AI summary
Various embodiments of the teachings herein include industrial network-based codeless tracking analytics methods for industrial software. Some methods include: identifying an industrial intranet topology through detection, and generating a configuration map of the industrial field according to the topology and/or an application program running list of the software; injecting a sidecar agent for codeless runtime tracking to an edge device of a target field device in the industrial field according to the configuration map, and performing data synchronization for the edge device and the sidecar agent corresponding to the edge device, at least one tracking point existing between the target edge device and the sidecar agent corresponding to the target edge device, and the target edge device running at least one application program of the industrial software; and obtaining analysis data from the tracking point when the target edge device runs an application program therein, to analyze the industrial software.