Packet Scheduler Reducing QoS Variance During Bandwidth Contention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network transmission systems face challenges in maintaining consistent quality of service (QoS) and throughput due to varying factors such as path length, flow control algorithms, and congestion avoidance, leading to imbalances during bandwidth contention, where well-performing connections may cannibalize resources from under-performing ones.
Innovation Solution
A method and system for multi-connection aware packet scheduling that measures relative QoS gains or penalties across connections, scheduling packets based on ratios of these penalties to prevent resource cannibalization, ensuring fair distribution and recovery for under-performing connections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If standard packet scheduling is used without QoS awareness, then device complexity is reduced, but QoS variance between connections increases and well-performing connections cannibalize throughput from under-performing ones
Solution Approach 1:
The patent implements feedback mechanisms where the scheduler continuously monitors QoS metrics (throughput, latency, packet loss) for each connection and adjusts scheduling decisions based on this feedback. This allows the system to dynamically respond to varying connection conditions and prevent throughput cannibalization while maintaining manageable complexity through adaptive rather than purely reactive control
Solution Approach 2:
The system changes scheduling parameters (such as weight factors, priority levels, or allocation ratios) based on measured QoS conditions of different connections. By dynamically adjusting these parameters rather than using fixed scheduling rules, the system achieves QoS consistency without requiring fundamentally complex scheduling architecture
2Reliability
If QoS-aware packet scheduling is implemented to prevent throughput cannibalization, then QoS consistency improves, but device complexity increases due to monitoring and ratio-based scheduling requirements
Solution Approach 1:
The scheduling system performs self-service by automatically monitoring its own QoS metrics and making autonomous scheduling decisions based on observed conditions. Each connection essentially schedules itself through the penalty ratio mechanism, where the scheduler simply responds to measured conditions rather than requiring complex centralized control logic, thereby achieving QoS consistency with moderate complexity
Data Source
AI summary
The systems and methods discussed herein provide for relative QoS gain measurement, and reduction in variance of QoS during bandwidth contention. In some implementations, the system measures relative QoS gains of, or penalties to, similar connections or network flows due to different factors that affect the transmission throughput and QoS. The system provides scheduling of packets on different connections or flows according to ratios of QoS penalties or gains on the connections or flows, preventing a well-performing connection or flow from cannibalizing throughput from under-performing connections or flows.