Modular Underlay Overlay Assurance via KPF Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current underlay/overlay assurance solutions are limited in their ability to provide flexible, broad, multiple overlay assurance over the underlay, are typically focused on a single type of underlay, and lack integration with third-party assurance products, as well as being unaware of how problems caused by one overlay or underlay can be mediated by another, leading to scalability issues and complexity.
Innovation Solution
A system and method that obtain and analyze telemetry data from multiple overlays and underlays using a Key Performance Factor model, perform Root Cause Analysis, and map actions with priorities to remediate issues, utilizing remediation modules that can enact changes in both underlays and overlays, and communicate with third-party systems, while leveraging AI and normalization techniques to ensure customer application satisfaction and optimize network cost.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If current underlay/overlay assurance solutions are tailored to one or two specific overlays, then they can provide detailed assurance for those specific applications, but they cannot provide flexible, broad, multiple overlay assurance and lack scalability
Solution Approach 1:
The patent implements a universal underlay assurance platform that can assure multiple different overlay applications simultaneously. The system uses a common data model and assurance framework that works across diverse overlays (SD-WAN, 5G, cloud networks, etc.), eliminating the need for separate specialized software for each overlay type while maintaining comprehensive assurance capabilities.
Solution Approach 2:
The patent segments the assurance system into modular components: telemetry collectors, data normalization layers, analysis engines, and remediation modules. Each component is independent and can be configured for different overlay types, allowing the system to scale to multiple overlays without increasing overall system complexity.
2Adaptability or versatility
If underlay/overlay assurance is focused on a single type of underlay infrastructure, then the solution can be optimized for that specific infrastructure, but it cannot integrate with third-party assurance products or handle multiple underlay types
Solution Approach 1:
The patent introduces telemetry adaptors and data normalization layers as intermediary components between different underlay infrastructures and the core assurance engine. These intermediaries translate various underlay telemetry formats into a unified data model, enabling the system to work with multiple underlay types and third-party products without direct integration complexity.
3Reliability
If each overlay requires its own specialized software that is not aware of other overlays, then the software can be optimized for that specific overlay, but the system lacks awareness of how problems in one overlay can be mediated by another
Solution Approach 1:
The patent merges multiple overlay assurance capabilities into a single unified platform. The system collects telemetry from multiple overlays simultaneously and uses a centralized analysis engine that understands relationships between different overlays, enabling cross-overlay problem mediation while reducing overall software complexity through consolidation.
4Adaptability or versatility
If the underlay/overlay assurance implementation is hardcoded with a two-level overlay-underlay hierarchy, then the system can be simple to implement, but it lacks flexibility to work with complex multi-level hierarchies or federated models
Solution Approach 1:
The patent implements a dynamic hierarchical model where the system can adapt to different hierarchy depths and structures based on the specific deployment scenario. The architecture supports two-level, multi-level, and federated models through configurable parameters rather than hardcoded structures, maintaining implementation simplicity while providing flexibility.
5Productivity
If network automation reduces human intervention through closed-loop automation, then operational efficiency improves, but the system requires sophisticated telemetry analysis and root cause analysis capabilities
Solution Approach 1:
The patent implements closed-loop automation with continuous feedback mechanisms. The system collects telemetry, analyzes it in real-time, automatically remediates detected issues, and verifies resolution through continued monitoring. This feedback loop enables high automated productivity while managing analysis complexity through iterative processing and prioritization.
Data Source
AI summary
Systems and methods include obtaining overlay telemetry data from a plurality of overlays, wherein each overlay is an application and there is a corresponding telemetry adaptor for each overlay; obtaining underlay telemetry data from one or more underlays, wherein each underlay includes physical infrastructure for supporting one or more of network, compute, and store functions for the plurality of overlays; analyzing the overlay telemetry data and the underlay telemetry data via a Key Performance Factor (KPF) model that correlates one or more of the plurality of overlays and the one or more underlays together; responsive to an anomaly or a threshold crossing based on the KPF model, performing a Root Cause Analysis (RCA) to identify a root cause of the anomaly or the threshold crossing; and mapping one or more actions with associated priorities to the root cause utilizing one or more remediation modules each associated with an overlay.