Packet Scheduling Throughput via Lazy QoS Batching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing QoS processing in data communication networks causes overhead delay, especially during high-throughput periods, leading to a bottleneck that limits maximum packet per second communication rates.
Innovation Solution
Implementing 'lazy' QoS processing by applying it only to a subset of packets based on throughput credits, allowing packets that bypass QoS processing to be immediately forwarded with the same QoS settings as packets of the flow for which QoS processing is applied.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If QoS processing is applied to every packet, then QoS accuracy is improved, but overhead delay increases and throughput decreases
Solution Approach 1:
The patent applies QoS processing to only a subset of packets (sampling approach) rather than every packet. Throughput credits are accumulated and used to selectively process packets, achieving partial QoS enforcement that maintains acceptable accuracy while dramatically reducing overhead delay and increasing throughput during high-demand periods
Solution Approach 2:
The patent segments packets into different processing paths: packets with accumulated throughput credits undergo QoS processing and are placed in QoS queues, while packets without sufficient credits bypass processing and are placed in output transmission buffers. This segmentation allows the system to handle different packet subsets differently, resolving the contradiction between processing accuracy and throughput
2Reliability
If QoS processing is applied to every packet, then QoS enforcement is improved, but overhead delay scales with throughput increases
Solution Approach 1:
Instead of enforcing QoS on every packet, the system enforces QoS on a sampled subset of packets based on throughput credit accumulation. This partial enforcement maintains QoS reliability for critical flows while avoiding the proportional overhead delay that would occur with universal packet processing
Solution Approach 2:
The system performs preliminary action by accumulating throughput credits for packets before QoS processing decisions are made. This allows the system to pre-identify which packets should undergo QoS processing, reducing the need for real-time inspection of every packet and thereby reducing overhead delay
3Measurement precision
If QoS processing is applied to every packet, then QoS accuracy is maintained, but processing overhead becomes a bottleneck
Solution Approach 1:
The patent implements partial QoS processing by applying inspection and classification only to packets that have accumulated sufficient throughput credits. This reduces the number of packets requiring complex QoS processing, thereby reducing device complexity and processing overhead while maintaining acceptable QoS accuracy through selective enforcement
Data Source
AI summary
Described embodiments improve the performance of a computer network via selectively forwarding packets to bypass quality of service (QoS) processing, avoiding processing delays during critical periods of high demand, increasing throughput and efficiency may be increased by sacrificing a small amount of QoS accuracy. QoS processing may be applied to a subset of packets of a flow or connection, referred to herein as “lazy” processing or lazy byte batching. Packets that bypass QoS processing may be immediately forwarded with the same QoS settings as packets of the flow for which QoS processing is applied, resulting in tremendous overhead savings with only minimal decline in accuracy. In case of backlog, packets may be collected together into an aggregated or ‘uber’ packet, with QoS processing applied based on a virtual size of the aggregated packet.