Software-Defined Endpoint Tracking for Non-Standard Fabric Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current IPDT frameworks treat all endpoints uniformly, failing to account for non-standard behaviors of IoT/OT devices, leading to inefficiencies and the need for workarounds like L2 flooding, and lack scalability in applying different policies based on endpoint type.
Innovation Solution
A distributed Software Defined Tracking (SDT) architecture that extends endpoint onboarding and probing logic beyond individual network devices to a fabric orchestrator, using event-driven and protocol-based methods to adapt to various endpoint behaviors, including IoT/OT devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If local packet gleaning and local probing are used for IPDT, then endpoint tracking is maintained at individual network devices, but the system cannot adapt to non-standard endpoint behaviors and lacks scalability for different endpoint types
Solution Approach 1:
The patent segments the endpoint tracking functionality by introducing fabric-wide endpoint type classification that divides endpoints into standardized and non-standardized categories. This segmentation allows different tracking approaches to be applied to different endpoint types, with standardized endpoints using traditional local probing and non-standardized endpoints receiving alternative tracking methods, thereby resolving the contradiction between adaptability and complexity
Solution Approach 2:
The patent introduces an intermediary mechanism (fabric-wide endpoint type classification and identification) that mediates between the uniform local probing approach and the need for customized tracking for IoT/OT devices. This intermediary layer enables the system to identify endpoint types and apply appropriate tracking strategies without requiring complete redesign of the existing IPDT framework
2Reliability
If uniform endpoint tracking is applied to all devices, then implementation is simple, but non-standard endpoints like IoT/OT devices cannot be properly tracked
Solution Approach 1:
The patent applies local quality by enabling differentiated tracking behavior based on endpoint type classification. Standardized endpoints continue to use traditional ARP-based probing while non-standardized endpoints (IoT/OT devices) receive customized tracking approaches. This localized adaptation improves tracking reliability for diverse endpoints without requiring complete system-wide complexity
3Adaptability or versatility
If L2 flooding is used as a workaround for non-standard endpoints, then endpoint discovery is achieved, but network efficiency deteriorates due to excessive traffic
Solution Approach 1:
The patent extracts the endpoint discovery function from the generic L2 flooding approach and implements it through targeted fabric-wide classification and type-specific probing. Instead of flooding the entire network with broadcast traffic, the system identifies non-standard endpoints through fabric-wide classification and applies customized, efficient discovery methods only where needed, thereby maintaining adaptability while eliminating excessive bandwidth consumption
Data Source
AI summary
Techniques for leveraging a software defined tracking architecture for endpoint probing are described. An orchestrator of a network fabric receives probing rules associated with a particular endpoint type. The orchestrator receives a notification that an event occurred with respect to an and point connected to the fabric edge device, the notification indicates that the endpoint is of the particular endpoint type. Based at least in part on the endpoint being of the particular endpoint type, the orchestrator transmits instructions, based at least in part on the probing rules, regarding how to probe the endpoint, to the fabric edge device.