QoS Network Flow Control via Traffic Duplication Testing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In multi-access networking environments, accurately measuring Quality-of-Service (QoS) metrics on new links and detecting QoS violations is challenging, leading to potential performance degradation and service denial.

Innovation Solution

A control technique that duplicates traffic over a new link using a Finite-State-Machine (FSM) based approach to obtain realistic QoS telemetry, combined with dynamic flow prioritization to manage traffic and ensure QoS compliance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traffic is duplicated over a new link for QoS testing, then accurate QoS measurement is achieved, but network overhead increases

Engineering Contradiction:
ImproveQoS measurement accuracyVSAvoidnetwork overhead
Core Design Contradiction:
Measurement precisionVSLoss of energy

Solution Approach 1:

The system performs QoS testing by duplicating traffic over a new link before fully committing to using that link for production traffic. This preliminary action allows the network to evaluate whether the new link meets required QoS thresholds (such as latency, jitter, or packet loss) before routing actual user traffic through it, thereby avoiding premature link activation that would waste network resources.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts traffic duplication parameters based on network conditions and QoS requirements. By monitoring QoS metrics during the testing phase and comparing them against thresholds, the system can determine when to activate, deactivate, or adjust the degree of traffic duplication, thereby optimizing the balance between measurement accuracy and network overhead.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If multiple links are used for traffic distribution, then network reliability is improved, but link selection complexity increases

Engineering Contradiction:
Improvenetwork reliabilityVSAvoidlink selection complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system continuously monitors QoS metrics on active links and uses this feedback to dynamically adjust traffic distribution. When a link fails to meet QoS thresholds or experiences degradation, the system automatically reduces or stops traffic duplication on that link and redirects traffic through alternative links, thereby maintaining reliability while adapting to changing network conditions.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system implements dynamic link selection and traffic distribution that adapts to real-time network conditions. Rather than using static routing rules, the system continuously evaluates QoS metrics and adjusts traffic flow distribution across multiple links dynamically, allowing the network to optimize reliability while managing complexity through adaptive rather than predetermined link selection.

Inventive Principle:
Principle #15Dynamics

3Reliability

If QoS testing is performed on new links, then service quality is ensured, but testing overhead and potential service denial increase

Engineering Contradiction:
Improveservice qualityVSAvoidtesting overhead
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system adjusts QoS testing parameters such as traffic duplication rate, testing duration, and threshold values based on network conditions and service requirements. By dynamically tuning these parameters, the system can perform adequate QoS validation while minimizing the impact on normal traffic flow and reducing testing overhead time.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system performs partial QoS testing by duplicating only a portion of traffic (rather than all traffic) over the new link during the testing phase. This partial action approach provides sufficient QoS measurement data to evaluate link suitability while minimizing the impact on overall network performance and reducing testing overhead.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4531363A1Quality-of-service network flow control
Publication Date: 2025.04.02 INTEL CORP
  • EP4531363A1 patent drawingFigure 1
  • EP4531363A1 patent drawingFigure 2
  • EP4531363A1 patent drawingFigure 3

AI summary

Quality of Service (QoS) network flow control may be accomplished through several systems and techniques. A device, when transmitting traffic over a first link, may copy that traffic to a second link to test the conformance with QoS requirements of the second link. After receiving in-band telemetry of the copied traffic on the second link that indicates that the second link meets the QoS requirements of the traffic, at least part of the traffic is transmitted by the second link.