Dynamic Queue Scheduling for Network Congestion Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Weight-based packet schedulers in network devices fail to effectively manage packet queuing and dequeuing across service cards, leading to congestion in downstream queues due to varying packet sizes and processing times, resulting in inefficient handling of packets.

Innovation Solution

A method where the current length of a downstream queue determines a scheduling value, which is sent to an upstream queue, allowing for dynamic adjustment of dequeuing based on the queue's state, preventing congestion by optimizing the scheduling weight.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If weight-based packet schedulers are used with fixed weights, then scheduling simplicity is maintained, but downstream queue congestion occurs due to varying packet sizes and processing times

Engineering Contradiction:

Improvescheduling simplicityVSAvoidqueue congestion prevention

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent implements dynamic weight adjustment where the weight for each traffic class is no longer fixed but changes based on the current state of downstream queues. The upstream service card continuously monitors downstream queue lengths and adjusts scheduling weights accordingly, transforming the static weight-based scheduler into a dynamic system that adapts to varying traffic conditions, packet sizes, and processing times across different service cards

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system establishes a feedback loop where the upstream service card receives information about downstream queue states (queue lengths, congestion levels) and uses this feedback to adjust its scheduling decisions. This closed-loop control mechanism allows the scheduler to respond to actual traffic conditions and downstream processing capabilities, preventing congestion by reducing upstream scheduling weight when downstream queues are full and increasing it when downstream queues have capacity

Inventive Principle:

Principle #23Feedback

2Stability of the object's composition

If upstream service card schedules packets based on fixed weights, then scheduling consistency is maintained, but downstream queues experience overflow when packet sizes vary due to compression/decompression or encryption/decryption

Engineering Contradiction:

Improvescheduling consistencyVSAvoiddownstream queue capacity

Core Design Contradiction:

Stability of the object's compositionVSQuantity of substance

Solution Approach 1:

The patent transforms the static weight-based scheduling into a dynamic system where weights are continuously adjusted based on downstream queue states. This allows the system to maintain scheduling consistency through a structured adjustment mechanism while adapting to varying packet quantities caused by compression, decompression, encryption, or decryption operations at different service cards

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The upstream service card implements feedback-based weight adjustment by monitoring downstream queue lengths and using this information to dynamically modify scheduling weights. This feedback mechanism ensures that the upstream card schedules packets at rates that match downstream processing capabilities, preventing queue overflow even when packet sizes vary due to processing operations

Inventive Principle:

Principle #23Feedback

3Productivity

If upstream service card processes packets at constant rate, then processing efficiency is maintained, but downstream queues become empty while upstream continues scheduling for other queues

Engineering Contradiction:

Improveprocessing efficiencyVSAvoidqueue state adaptability

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamic weight adjustment where the scheduling weight for each traffic class changes based on downstream queue states. This allows the upstream service card to maintain high processing efficiency by continuously adapting its scheduling decisions to current downstream conditions, directing more packets to queues that have capacity and fewer to queues that are full, thereby optimizing overall system throughput

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system uses feedback from downstream queue states to dynamically adjust upstream scheduling weights. This feedback mechanism enables the upstream service card to respond to changes in downstream processing rates and queue conditions, maintaining processing efficiency by aligning upstream packet generation with downstream consumption rates for each traffic class

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS7899068B1Coordinated queuing between upstream and downstream queues in a network device

Publication Date: 2011.03.01 JUNIPER NETWORKS INC

AI summary

A system determines a scheduling value based on a current length of a downstream queue in a network device. The system sends the scheduling value from the downstream queue to an upstream queue and schedules dequeuing of one or more data units, destined for the downstream queue, from the upstream queue based on the scheduling value.