Adaptive SD-WAN Probing for QoE Accuracy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Software-defined WAN (SD-WAN) edge network devices face resource constraints, limiting their ability to continuously monitor all possible data paths for packet flows, and making it difficult to determine which specific data paths to prioritize with a limited probe packet budget.

Innovation Solution

The techniques involve adaptive application-specific probing based on historical models of application traffic, where probe packets are sent only to servers historically more likely to be used for meaningful flows, determined by analyzing Quality of Experience (QoE) metrics and using machine learning to predict data flow patterns.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If SD-WAN edge network devices continuously monitor all possible data paths, then measurement precision of QoE metrics is improved, but device complexity and resource consumption increase beyond available computational power

Engineering Contradiction:
ImproveQoE metrics collection accuracyVSAvoidmonitoring complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the monitoring task by dividing all possible data paths into two categories: high-probability paths (to be monitored) and low-probability paths (not monitored). This segmentation is achieved by analyzing historical flow data to identify which servers are most likely to be accessed, thereby focusing monitoring resources on the most critical paths while accepting that some paths will not be monitored.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by differentiating the monitoring approach based on the specific characteristics of each data path. Instead of uniform monitoring, the system monitors only those data paths to servers that have historical evidence of being accessed, thereby applying monitoring intensity locally where it matters most rather than uniformly across all paths.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If SD-WAN edge network devices send probe packets to all possible servers, then measurement precision of QoE metrics is improved, but the probe packet budget is exceeded

Engineering Contradiction:
ImproveQoE metrics collection accuracyVSAvoidprobe packet count
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent segments the set of all possible servers into two groups: servers that are historically likely to be accessed (selected for probing) and servers that are not historically accessed (excluded from probing). This segmentation allows the system to send probe packets only to the necessary subset of servers, thereby staying within the probe packet budget while maintaining measurement precision for the most important paths.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by sending probe packets only to a subset of servers rather than all possible servers. The system determines the optimal subset size and composition based on historical flow data, ensuring that the probe packet count remains within the budget while still providing sufficient measurement accuracy for the most critical data paths.

Inventive Principle:
Principle #16Partial or excessive action

3Ease of operation

If SD-WAN edge network devices use static probe configurations, then ease of operation is improved, but adaptability to changing data flow patterns deteriorates

Engineering Contradiction:
Improveprobe configuration simplicityVSAvoidadaptation to flow changes
Core Design Contradiction:
Ease of operationVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamics by transitioning from static probe configurations to dynamic, adaptive configurations. The system continuously collects and analyzes historical flow data to identify changing patterns in server access, and automatically adjusts the probe packet destination list accordingly. This dynamic adaptation allows the system to respond to evolving network usage patterns without manual reconfiguration.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent applies feedback by using historical flow data as feedback to continuously improve the probe packet destination selection. The system analyzes actual server access patterns and uses this feedback to refine the list of probed servers over time, thereby adapting to changing network conditions while maintaining operational simplicity through automated adjustment.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS20250193103A1Adaptable software defined wide area network application-specific probing
Publication Date: 2025.06.12 JUNIPER NETWORKS INC
  • US20250193103A1 patent drawing
  • US20250193103A1 patent drawing
  • US20250193103A1 patent drawing

AI summary

Systems, devices and techniques for an adaptive application-specific probing scheme are disclosed. An example network device includes memory configured to store a network address and probe protocol usable for probing a first network device associated with a source of an application, and one or more processors configured to determine a network address and probe protocol usable for probing the first network device, wherein the first network device comprises a server that is responsive to the probing, the server executing the application for the data flow, or a closest network device, to the server, that is responsive to the probing. The one or more processors are also configured to send to a second network device at a location serviced by the application, a message specifying the network address and probe protocol usable for probing the first network device.