Network Device Cut-Through Processing Egress Congestion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional network devices using cut-through processing face delays due to traffic congestion at egress ports, leading to increased latency and potential packet drops, as they often require queuing packets in a buffer before transmission, especially when the egress port is congested.
Innovation Solution
A network device determines the congestion state of the egress port and selectively uses either cut-through processing by transmitting packets directly from a transient buffer or store-and-forward processing by queuing packets, based on the congestion state, to avoid additional latency and packet drops.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If cut-through processing is used to reduce latency, then transmission speed is improved, but packet drops increase due to egress port congestion
Solution Approach 1:
The patent implements dynamic path selection that adapts to changing network conditions. The system monitors egress port congestion states and dynamically switches between cut-through processing (for non-congested ports) and store-and-forward processing (for congested ports), allowing the network device to optimize both speed and reliability based on real-time conditions
Solution Approach 2:
The patent changes the processing mode parameter based on congestion state. When egress port congestion is detected, the system transitions from cut-through mode to store-and-forward mode, effectively changing the operational parameter to prevent packet drops while maintaining low latency when conditions permit
2Reliability
If store-and-forward processing is used to ensure reliable packet delivery, then packet drops are reduced, but transmission latency increases
Solution Approach 1:
The system dynamically selects between store-and-forward and cut-through processing based on real-time congestion monitoring. When egress ports are non-congested, the system switches to cut-through mode to minimize latency, while only using store-and-forward mode when congestion is detected, thus reducing the time penalty associated with reliable delivery
3Reliability
If packets are queued in buffer before transmission, then egress port congestion is managed, but additional latency is introduced
Solution Approach 1:
The patent extracts the queuing operation from the mandatory packet path by implementing selective cut-through processing. Instead of always queuing packets before transmission, the system extracts the queuing step entirely for packets destined to non-congested egress ports, allowing direct transmission from ingress to egress while maintaining queuing only when necessary for congestion management
Data Source
AI summary
A method for processing network packets in a network device is described. A network packet is stored in a transient buffer as the network packet is being received at an ingress port of the network device. After at least a first portion of the network packet has been received and before the entire network packet has been received: the first portion is processed to identify an egress port of the network device from which the network packet is to be transmitted; a congestion state of the egress port is determined; and the network packet is selectively transferred from the transient buffer to the identified egress port for transmission from the network device or a different action is performed on the network packet, based on the congestion state.