On-Demand SD-WAN Probing for Scalable QoE Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing SD-WAN systems face challenges in efficiently managing network performance metrics due to the increasing variety and number of subscriber sessions, leading to inefficiencies in resource consumption and overhead, particularly in generating Quality of Experience (QoE) metrics across multiple links and queues.
Innovation Solution
Implementing on-demand active synthetic probing techniques in SD-WAN appliances that initiate probing only on active links and queues, using application-specific probe packets to reduce unnecessary bandwidth consumption and processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If preconfigured probe packets are sent over every WAN link continuously, then QoE metrics are always available for all links, but bandwidth consumption and processing load increase significantly
Solution Approach 1:
The patent implements dynamic probing by transitioning from static continuous probing of all links to dynamic on-demand probing. The system actively monitors data flow presence and dynamically adjusts probing behavior - sending probe packets only when data flows are detected on specific links, and stopping probes when flows cease. This dynamic adaptation resolves the contradiction by making measurement availability responsive to actual network conditions rather than continuously maintaining probes regardless of traffic presence.
Solution Approach 2:
The patent employs periodic monitoring of data flow presence to determine when to initiate or cease probing. Instead of continuous probing, the system performs periodic checks to detect data flow conditions and adjusts probing accordingly. This periodic action allows the system to maintain QoE metrics availability when needed while reducing bandwidth consumption during idle periods, resolving the contradiction between measurement precision and energy loss.
2Measurement precision
If probe packets are sent on all links simultaneously, then complete QoE data is collected for link selection, but computing resources and processing load increase
Solution Approach 1:
The patent applies local quality by tailoring probing behavior to specific links based on local conditions (data flow presence). Instead of uniformly probing all links with the same intensity, the system adjusts probing activity locally - actively probing links with data flows and passively or ceasing probes on links without flows. This localized adaptation maintains QoE data completeness for active links while reducing overall processing load, resolving the contradiction between measurement precision and device complexity.
Solution Approach 2:
The patent implements partial action by probing only the subset of links that currently carry data flows, rather than comprehensively probing all available links. This selective partial probing provides sufficient QoE data for link selection decisions while significantly reducing computing resources and processing load compared to full comprehensive probing, thus resolving the contradiction between measurement precision and device complexity.
3Reliability
If continuous probing is performed on all links, then link performance is always monitored, but scalability is limited due to overhead
Solution Approach 1:
The patent implements dynamic probing that adapts to network conditions, allowing the system to scale effectively. By monitoring data flow presence and dynamically adjusting probing activity, the system can accommodate increasing numbers of links and flows without linearly increasing overhead. The dynamic nature enables the system to maintain reliable performance monitoring for active links while automatically reducing monitoring intensity for inactive links, thus improving scalability without sacrificing reliability.
Solution Approach 2:
The system performs self-service by automatically detecting data flow presence and autonomously determining when to initiate or cease probing without external intervention. This self-service capability reduces the need for manual configuration and simplifies scaling operations. The system self-regulates probing activity based on observed traffic patterns, enabling reliable link performance monitoring to be maintained as the network scales while reducing the operational overhead associated with manual management.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In general, the disclosure describes techniques for evaluating application quality of experience metrics over a software-defined wide area network. For instance, a network device may receive an application data packet of a data flow for an application. In response to receiving the application data packet, the network device may assign the data flow to a first link of a plurality of links and initiate a probing process for the data flow on the first link to determine one or more quality of experience (QoE) metrics for the first link. The network device may, at a later time, detect that the data flow is no longer being received. In response to detecting that the data flow is no longer being received, the network device may cease the probing process for the data flow on the first link.