Aggregated Signal Feedback for SaaS Root Cause Detection in Multi-Cloud SD-WAN
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the root cause of degraded user experience in SaaS applications within multi-cloud SD-WAN deployments are inadequate, as they rely on granular telemetry metrics that do not accurately reflect application performance and often misattribute issues to network paths rather than local edge conditions.
Innovation Solution
An Aggregated Signal Feedback (ASF) mechanism is introduced, where edge devices collect telemetry data, detect anomalies using machine learning, and provide an indication to SaaS providers to determine the root cause of performance degradation, incorporating local edge metrics such as CPU usage, memory, and queue depth.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If granular telemetry metrics are used to determine root cause of degraded user experience, then detailed network path information is obtained, but measurement precision deteriorates due to misattribution of issues to network paths rather than local edge conditions
Solution Approach 1:
The patent segments the telemetry data into two distinct categories: granular network path metrics (latency, loss, jitter) and local edge device metrics (CPU usage, memory, queue depth). This segmentation allows the system to analyze both types of data separately and combine them for accurate root cause determination, preventing misattribution of edge device issues to network path problems.
Solution Approach 2:
The patent introduces an intermediary analysis layer that correlates network path metrics with local edge device metrics. This intermediary mechanism acts as a mediator between network telemetry and application performance data, enabling accurate attribution of performance degradation to the correct source (network path vs. edge device) by comparing timing and correlation of anomalies across both metric types.
2Reliability
If multiple telemetry metrics are collected from edge devices, then comprehensive monitoring is achieved, but device complexity increases due to data collection and anomaly detection requirements
Solution Approach 1:
The patent implements partial action by collecting only the most critical telemetry metrics from edge devices (CPU usage, memory utilization, queue depth) rather than attempting to monitor all possible device parameters. This selective approach provides sufficient reliability for root cause determination while keeping the data collection and processing complexity at edge devices manageable.
Solution Approach 2:
The patent enables edge devices to perform self-service anomaly detection by comparing their local metrics against predefined thresholds and patterns. The edge devices autonomously identify and report anomalies without requiring complex centralized analysis, reducing the processing burden on individual devices while maintaining comprehensive monitoring capability.
3Measurement precision
If local edge metrics are incorporated into anomaly detection, then root cause identification is improved, but ease of operation deteriorates due to more complex data analysis requirements
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously monitors both network path metrics and local edge device metrics, automatically correlates anomalies between them, and provides actionable insights about root causes. This feedback loop simplifies operation by automatically performing the complex analysis that would otherwise require manual intervention, presenting operators with clear conclusions about whether performance degradation originates from network paths or edge devices.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
In embodiments, an edge device located at an edge of a local network provides connectivity between the local network and a cloud-based software as a service (SaaS) provider via one or more interfaces. The edge device obtains telemetry data associated with the edge device for a plurality of metrics. The edge device makes a determination that one or more of the plurality of metrics is anomalous. The edge device sends, based on the determination, an indication of the determination to the SaaS provider. The SaaS provider uses the indication to determine a root cause of an application served by the SaaS provider experiencing degraded application performance.