Traffic Shaping Scheduling via Flow Group Queue Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing traffic shaping scheduling methods, such as Subscriber Group Queue shaping, fail to effectively manage traffic for individual services within a subscriber group, leading to bandwidth blockage issues when a single service consumes excessive bandwidth, thereby affecting other services.

Innovation Solution

A traffic shaping scheduling method that assigns a unique Subscriber Queue identifier to each subscriber, with multiple Flow Queues corresponding to services configured into separate Flow Group Queues, using token buckets to limit data traffic, allowing each service's traffic to be managed independently to prevent bandwidth blockage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If GQ shaping method is used to limit total traffic of subscriber group, then total bandwidth utilization is improved, but individual service traffic control capability deteriorates

Engineering Contradiction:
Improvetotal bandwidth utilizationVSAvoidtraffic control granularity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the GQ traffic control into multiple FQs, each corresponding to a specific service. By dividing the GQ into multiple FQs with separate token buckets, the system achieves fine-grained traffic control for each service while maintaining overall GQ bandwidth management, thus resolving the contradiction between total bandwidth utilization and traffic control granularity.

Inventive Principle:
Principle #1Segmentation

2Productivity

If total traffic of GQ is limited, then bandwidth allocation among multiple SQs is improved, but single service bandwidth occupation control deteriorates

Engineering Contradiction:
Improvebandwidth allocation efficiencyVSAvoidbandwidth blockage
Core Design Contradiction:
ProductivityVSObject-affected harmful factors

Solution Approach 1:

The patent segments the GQ token bucket into multiple FQ-specific token buckets. Each FQ has its own token bucket that independently limits its service traffic, preventing any single service from monopolizing bandwidth. This segmentation allows bandwidth allocation among multiple SQs while simultaneously controlling single service bandwidth occupation, resolving the contradiction between bandwidth allocation efficiency and bandwidth blockage prevention.

Inventive Principle:
Principle #1Segmentation

3Device complexity

If service traffic is not independently controlled, then system complexity is reduced, but service quality assurance deteriorates

Engineering Contradiction:
Improvescheduling system complexityVSAvoidservice quality assurance
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent segments the scheduling system into multiple FQs, each with independent token buckets and scheduling mechanisms. This segmentation enables independent service traffic control, ensuring service quality for each FQ while maintaining manageable system complexity through modular design. Each FQ can be configured and controlled independently, allowing service quality assurance without excessive system complexity.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP2357764B1Traffic shaping scheduling method, traffic shaping scheduling apparatus, and routing device
Publication Date: 2017.08.23 HUAWEI TECH CO LTD
  • EP2357764B1 patent drawingFigure 1
  • EP2357764B1 patent drawingFigure 2
  • EP2357764B1 patent drawingFigure 3~4

AI summary

A traffic shaping scheduling method, a traffic shaping scheduling apparatus, and a routing device are provided. The method includes: obtaining a Subscriber Queue (SQ) Identifier (ID) (SQID) of an SQ passing previous-level scheduling, querying at least one Flow Group Queue (FGQ) ID (FGQID) corresponding to the SQID, in which each FGQID corresponds to one or more Flow Queues (FQs) of the SQ; judging whether the corresponding FQ satisfies a scheduling condition according to the number of currently remaining tokens in a token bucket corresponding to each FGQID, and scheduling the FQs satisfying the scheduling condition out of the queue. Thus, each service of the SQ is respectively configured to the corresponding FGQ, and the traffic of each FGQ corresponding to each service is limited, so that bandwidth traffic occupied by each service is limited, thereby preventing a situation that other services of the subscriber or other subscribers are blocked due to the excessively large traffic of the single service of the subscriber.