SDN Controller Path Intersection Fault Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network monitoring mechanisms struggle to efficiently detect and localize anomalies in computer networks, often requiring extensive probe packets and manual troubleshooting, which can be resource-intensive and error-prone.
Innovation Solution
A Software-Defined Networking (SDN) controller implements techniques to detect anomalies by sending probe packets to measure Quality of Experience (QoE) metrics, evaluates traffic flows to construct network topology, and optimizes probing processes by pruning redundant paths, thereby reducing network traffic and processor usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If extensive probe packets are sent to detect anomalies in each link, then measurement precision is improved, but use of energy and processor usage increase
Solution Approach 1:
The network is segmented into multiple paths, and anomaly detection is performed path-by-path rather than link-by-link. The controller receives QoE metrics from probes traversing entire paths and determines anomalies at the path level, then localizes root causes by analyzing intersections between affected paths, avoiding the need to send probes to every individual link
Solution Approach 2:
A single probe packet traversing a path serves multiple functions: it measures QoE metrics for the entire path, detects anomalies along the path, and provides data for root cause localization when combined with results from other paths. This multi-functionality reduces the total number of probes needed compared to individual link monitoring
2Reliability
If probes are sent to each node to localize root cause, then reliability of fault localization is improved, but loss of time and productivity decrease
Solution Approach 1:
The controller pre-computes the intersection of multiple paths to identify potential root cause locations before anomalies occur. When anomalies are detected in multiple paths, the pre-computed intersection information enables immediate root cause localization without requiring additional probing or manual troubleshooting
Solution Approach 2:
The system uses feedback from QoE metrics of multiple paths to continuously refine anomaly detection and root cause localization. By analyzing patterns across multiple path measurements, the controller can reliably identify root causes and adapt to changing network conditions
3Measurement precision
If redundant probing processes are executed on all links, then measurement completeness is improved, but loss of substance (network traffic) increases
Solution Approach 1:
Multiple probing operations are merged into single probe packets that traverse entire paths covering multiple links. A single probe measuring path QoE metrics simultaneously provides measurement data for all links along that path, eliminating redundant probing and reducing overall network traffic while maintaining monitoring completeness
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The disclosure describes techniques for network monitoring and fault localization. For example, a controller comprises one or more processors operably coupled to a memory configured to: receive a first one or more Quality of Experience (QoE) metrics measured by a first probe traversing a first path comprising one or more links; receive a second one or more QoE metrics measured by a second probe traversing a second path comprising one or more links; determine, from the first one or more QoE metrics, that the first path has an anomaly; determine, from the second one or more QoE metrics, that the second path has an anomaly; and determine, in response to determining the first path and the second path has an anomaly, based on the type of metrics and the type of links, that an intersection between the first path and the second path is a root cause of the anomaly.