Dynamic Probe Parameter Synthesis for SD-WAN SLA Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional active probing methods for evaluating service level agreement (SLA) metrics in Software-Defined Wide Area Networks (SD-WANs) are inaccurate due to static parameter configuration, varying network conditions, and mismatch between probe packets and actual traffic patterns, leading to undesirable network behavior.
Innovation Solution
Collecting historical data to determine patterns in packet flows, assigning ranges, and sending active probe packets based on these patterns to mimic actual network traffic, thereby improving the accuracy of SLA metrics and link selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If static parameters are used for active probe configuration, then device complexity is reduced, but measurement precision deteriorates due to mismatch with actual traffic patterns
Solution Approach 1:
The patent transforms static probe parameters into dynamic parameters that adapt to changing network conditions. The system continuously monitors actual traffic patterns and adjusts probe parameters (packet size, interval, count) accordingly, allowing the probing mechanism to evolve from a fixed configuration to a adaptive system that mirrors real traffic behavior.
Solution Approach 2:
The system enables the probe configuration to self-adjust based on observed traffic patterns without requiring manual intervention. By automatically analyzing historical traffic data and synthesizing optimal probe parameters, the system makes the probing mechanism self-configuring, reducing operational complexity while maintaining high measurement precision.
2Measurement precision
If manual tuning of probe parameters is performed, then measurement precision improves, but ease of operation deteriorates due to time-consuming configuration
Solution Approach 1:
The system eliminates manual tuning by implementing automatic parameter synthesis based on historical traffic data analysis. The probe configuration process becomes self-service, where the system autonomously determines optimal parameters without operator intervention, thereby maintaining high measurement precision while dramatically improving ease of operation.
Solution Approach 2:
The system performs preliminary analysis of traffic patterns before configuring probes, synthesizing parameters in advance based on observed patterns. This preliminary action enables the system to have probe parameters ready optimized for current conditions without requiring manual tuning at the time of deployment or adjustment.
3Measurement precision
If probe packets are sent frequently to capture dynamic network conditions, then measurement precision improves, but use of energy increases due to higher probe volume
Solution Approach 1:
The system dynamically adjusts probe packet frequency based on actual network conditions and traffic patterns. Rather than sending probes at a fixed high rate, the system adapts the probing intensity to match the variability of real traffic, maintaining measurement precision while reducing unnecessary probe transmissions during stable periods.
Solution Approach 2:
The system changes probe parameters (interval, count, size) based on synthesized patterns from historical data. By adjusting these parameters dynamically, the system optimizes the balance between measurement accuracy and resource consumption, sending more probes when conditions warrant and fewer when the network is stable.
4Ease of operation
If static SLA parameters are used for path selection, then ease of operation improves, but adaptability deteriorates as network conditions change
Solution Approach 1:
The system transforms static SLA parameters into dynamic parameters that adapt to changing network conditions. By continuously monitoring actual performance and traffic patterns, the system adjusts SLA thresholds and path selection criteria in real-time, maintaining ease of operation through automated adaptation while significantly improving link selection adaptability.
Data Source
AI summary
An example network device includes a memory configured to store a plurality of counts of packets of a data flow. The network device also includes one or more processors in communication with the memory. The one or more processors are configured to determine the plurality of counts of packets of the data flow, wherein each count of the plurality of counts includes a number of packets occurring in a predetermined time period. The one or more processors are configured to assign a corresponding range to each count of the plurality of counts, so as to assign a plurality of corresponding ranges. The one or more processors are also configured to determine a pattern in the plurality of corresponding ranges and send a number of active probe packets based on the determined pattern.