Sub-flow Queue Management for Real-Time Service Fairness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data queuing management techniques in communication systems with varying bandwidth conditions lead to increased queuing delay and delay variance, particularly affecting real-time services.
Innovation Solution
A method and node for sub-flow based queuing management of Protocol Data Units (PDUs), where each PDU is associated with a flow reference based on header information, grouped into sub-flows, and prioritized based on sub-flow length, with PDUs being selected and sent according to priority indicators, allowing for reduced delay and delay variance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing data queuing management techniques are used in systems with varying bandwidth conditions, then the system can handle variable link capacity, but the delay and delay variance for real-time services increase
Solution Approach 1:
The patent segments the queue into multiple sub-queues based on service types (real-time services, bandwidth-greedy services, background services). Each sub-queue is managed independently with service-specific parameters, allowing the system to adapt to varying bandwidth conditions while maintaining low delay for real-time services by preventing them from being blocked by other service types.
2Adaptability or versatility
If existing data queuing management techniques are used in systems with varying bandwidth conditions, then the system can handle variable link capacity, but the delay variance for real-time services increases
Solution Approach 1:
The patent applies different queuing parameters and management strategies to different sub-queues based on their specific service requirements. Real-time service sub-queues use parameters optimized for low delay and delay variance, while other sub-queues can use parameters optimized for their respective service types, thereby stabilizing delay variance for real-time services regardless of overall system bandwidth variations.
3Device complexity
If a single queue is used for all services, then the device complexity is low, but the reliability of real-time services deteriorates under varying bandwidth conditions
Solution Approach 1:
The patent divides the single queue into multiple service-type-specific sub-queues, segmenting traffic based on service characteristics. This segmentation improves real-time service reliability by ensuring their packets are not blocked by bandwidth-greedy or background services, while the overall structure remains manageable through automated service type detection and assignment.
4Reliability
If service-type-specific queues are implemented, then the reliability and performance of real-time services improve, but the device complexity increases
Solution Approach 1:
The patent implements mechanisms where packets automatically identify their service type and are assigned to appropriate sub-queues without complex manual configuration. The system self-adjusts by detecting service characteristics and applying appropriate queuing parameters, reducing the operational complexity despite the multi-queue structure.
Data Source
Figure 1~3
Figure 2~5
Figure 4
AI summary
This disclosure relates to a node (50) and a method therein. By grouping (108, 206, 410) PDUs into sub flows based on information elements of the PDUs and selecting (110, 210, 416) PDUs from sub flow according to a priority indicator calculated for each sub flow, the selected PDUs are sent (112, 212, 420) in an order that provides a fairer resource sharing for real time like service sessions at the expense of bandwidth greedy applications. The present disclosure provides reduced delay variability and a reduced PDU loss probability for real time like services when network transport resources are shared with bandwidth greedy services, such as file transfers over TCP.