Bonded Channel Packet Scheduling for DOCSIS Traffic Fairness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing packet scheduling and queuing systems in DOCSIS networks face challenges in ensuring fairness and efficient traffic distribution between different generations of modems, particularly between DOCSIS 3.1 and DOCSIS 4.0, leading to potential service disruptions and uneven queuing due to varying channel bandwidths and modulation schemes.
Innovation Solution
A dynamically weighted shortest-queue-first (DW-SQF) algorithm that considers channel rate, quality-of-service (QoS) factors, and saturation thresholds to dynamically adjust queue occupancies, ensuring fair allocation and scheduling of packets across bonded channels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional packet scheduling is used in DOCSIS networks, then existing traffic flow is maintained, but network traffic fairness between different modem generations deteriorates
Solution Approach 1:
The patent implements a dynamically weighted shortest-queue-first algorithm where queue weights are adjusted in real-time based on channel occupancy thresholds. The system transitions from static scheduling to dynamic scheduling that adapts to changing network conditions, particularly balancing traffic between DOCSIS 3.1 and 4.0 modems as channels become saturated. This resolves the contradiction by making the scheduling mechanism flexible enough to maintain fairness while adapting to varying traffic loads.
Solution Approach 2:
The system continuously monitors channel occupancy levels and uses this feedback to adjust scheduling decisions. When dominant channels (DOCSIS 4.0) reach their saturation threshold, the algorithm automatically shifts traffic to remaining channels (DOCSIS 3.1), ensuring fair bandwidth distribution. This feedback mechanism enables the system to maintain traffic fairness while efficiently utilizing available network capacity across different modem generations.
2Productivity
If dominant channels are prioritized for DOCSIS 4.0 modems, then throughput for new generation modems is improved, but service continuity for legacy modems deteriorates
Solution Approach 1:
The system pre-establishes saturation thresholds for dominant channels and configures the weighted shortest-queue-first algorithm in advance to handle channel saturation scenarios. By preparing the scheduling mechanism beforehand with predefined threshold-based weight adjustments, the system ensures that when DOCSIS 4.0 channels become saturated, traffic is automatically and seamlessly redirected to DOCSIS 3.1 channels, maintaining service continuity without disruption.
Solution Approach 2:
The algorithm dynamically changes the weight parameters assigned to different queues based on channel occupancy. When dominant channels exceed their saturation threshold, the system modifies the weight parameters to prioritize remaining channels, thereby adjusting traffic distribution parameters in real-time. This parameter change mechanism allows the system to maximize throughput for DOCSIS 4.0 modems while guaranteeing service continuity for DOCSIS 3.1 modems.
3Reliability
If complex scheduling algorithms are implemented to achieve fairness, then traffic distribution fairness is improved, but device complexity increases
Solution Approach 1:
The patent applies different scheduling weights to different queues based on their specific channel characteristics and occupancy levels. Rather than using a single complex algorithm for all traffic, the system assigns local quality parameters (weights) to specific queues, where dominant channel queues receive different weights than remaining channel queues. This localized approach achieves traffic fairness through simple, targeted weight adjustments rather than complex global optimization.
Solution Approach 2:
The system achieves traffic fairness by changing simple numerical parameters (queue weights) rather than implementing complex scheduling logic. The weighted shortest-queue-first algorithm modifies only the weight parameter based on channel saturation status, keeping the underlying scheduling mechanism simple while achieving fair traffic distribution. This parameter-based approach maintains low device complexity while improving traffic fairness.
Data Source
AI summary
In an embodiment, a method includes receiving a data packet with a quality-of-service (QoS) factor to be transmitted to an endpoint device, identifying a bonding group including multiple channels of dominant and remaining channels with each channel having respective queues, and determining queue occupancies for queues associated with the channels. The queues correspond to the QoS factor. The method also includes retrieving weights corresponding the QoS factor that are associated with the channels from a weight table. The weights are calculated based at least in part on a saturation threshold of the dominant channels. The method further includes generating adjusted queue occupancies based on the queue occupancies and the weights, selecting a first queue associated with a first channel based on the adjusted queue occupancies, and scheduling the first data packet for transmitting to the endpoint device via the first queue.