Centralized Probe Allocation for SD-WAN Link QoE Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In Software-Defined Wide Area Networks (SD-WANs), the current methods for determining Quality of Experience (QoE) metrics through probe packets consume excessive bandwidth and resources, particularly when measuring complex paths, due to redundant probing across multiple links, which complicates network performance optimization.
Innovation Solution
A centralized SDN controller constructs a network topology map, selects a single node to measure QoE metrics for each link, and extrapolates metrics for reverse directions and complex paths, reducing redundant probing and optimizing resource usage by distributing probing responsibilities uniformly across nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If each node device transmits probe packets over every connected link to measure QoE metrics, then measurement precision is improved, but bandwidth consumption and resource usage increase significantly
Solution Approach 1:
The patent segments the probing task by dividing the network into regions or zones, where only selected boundary nodes perform probing for specific links. This segmentation reduces redundant probing while maintaining comprehensive coverage, as each probe measures specific link characteristics that can be aggregated to infer overall network QoE metrics.
Solution Approach 2:
The patent makes probe packets multi-functional by designing them to simultaneously measure multiple QoE parameters (latency, jitter, packet loss) across multiple links through strategic placement and routing. A single probe packet can serve multiple measurement purposes, reducing the total number of probes needed while maintaining measurement precision.
2Measurement precision
If node devices send probe packets over complex multi-link paths to determine QoE metrics, then measurement completeness is improved, but resource consumption (CPU, memory, power) increases
Solution Approach 1:
The patent applies preliminary action by pre-computing optimal probe paths and selecting which nodes should perform probing before actual measurement begins. The controller determines the minimum set of probes needed to infer complex path QoE metrics by combining measurements from simpler paths, eliminating the need for exhaustive probing of all complex paths and reducing node device power consumption.
Solution Approach 2:
The patent introduces a centralized controller as an intermediary that coordinates probing activities across the network. The controller collects raw measurements from selected nodes and computes derived QoE metrics for complex paths centrally, rather than requiring each node to independently perform exhaustive probing. This mediator approach distributes computational burden and reduces individual node power consumption.
3Reliability
If multiple nodes perform redundant probing processes on the same link, then measurement reliability is improved, but device complexity and operational complexity increase
Solution Approach 1:
The patent implements feedback mechanisms where the centralized controller monitors probe results and dynamically adjusts the probing plan. When sufficient measurement reliability is achieved through aggregated data from multiple nodes, the controller reduces redundant probing. The feedback loop ensures reliability requirements are met while automatically eliminating unnecessary complexity in the probing plan.
Solution Approach 2:
The patent changes the parameter of probing frequency and participation by node dynamically based on network conditions and measurement needs. Instead of fixed redundant probing by all nodes, the system adjusts which nodes probe which links based on current reliability requirements, reducing operational complexity while maintaining measurement reliability through adaptive parameter changes.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In general, the disclosure describes techniques for measuring edge-based quality of experience (QoE) metrics. For instance, a network device may construct a topological representation of a network, including indications of nodes and links connecting the nodes within the network. For each of the links, the network device may select a node device of the two node devices connected by the respective link to measure one or more QoE metrics for the respective link, with the non-selected node device not measuring the QoE metrics. In response to selecting the selected node device, the network device may receive a set of one or more QoE metrics for the respective link for data flows flowing from the selected node device to the non-selected node device. The network device may store the QoE metrics and determine counter QoE metrics for data flows flowing from the non-selected node device to the selected node device.