Network Device Data Packet Forwarding Latency Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The single-component data packet forwarding architecture in network devices introduces latency due to its inability to scale with the number of ingress network interfaces, as it often buffers all arriving data packets, leading to inefficiencies in forwarding data packets.
Innovation Solution
The architecture is segregated into two components: a collision detector and forwarder (CDF) and a data packet buffer (DPB), where the CDF selects data packets based on priority policies to forward them while the DPB buffers or discards packets, optimizing forwarding and reducing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single-component architecture is used for data packet forwarding, then the device complexity is reduced, but the latency increases due to inability to scale with ingress network interfaces and excessive buffering
Solution Approach 1:
The patent divides the single forwarding component into two separate components: a collision detector and forwarder (CDF) and a data packet buffer (DPB). The CDF handles packet reception, collision detection, and forwarding decisions, while the DPB handles selective buffering. This segmentation allows each component to specialize in specific functions, reducing latency for non-colliding packets while maintaining scalability with multiple ingress interfaces.
2Reliability
If all arriving data packets are buffered in a single component, then packet loss is reduced, but the latency increases and scalability is limited
Solution Approach 1:
The patent implements selective buffering where only colliding data packets are buffered in the DPB, while non-colliding packets are forwarded immediately by the CDF without buffering. This local quality approach ensures that buffering resources are applied only where necessary (for colliding packets), maintaining reliability for packet delivery while minimizing latency for the majority of non-colliding packets.
3Device complexity
If the single component buffers all packets, then packet collision handling is simplified, but the output bandwidth is reduced and scalability is limited
Solution Approach 1:
The patent extracts the buffering function from the forwarding component, creating a separate DPB dedicated solely to buffering colliding packets. This allows the CDF to focus on high-speed forwarding operations with minimal control logic, while the DPB handles buffer management independently. The extraction enables parallel operation of forwarding and buffering, increasing output bandwidth and scalability.
Data Source
AI summary
A method and network device for forwarding data packets. Specifically, the method and network device disclosed herein separate the known data packet forwarding architecture in network devices, often implemented using a single component, into two components. In implementing the pair of components, functionalities directed to forwarding data packets versus buffering data packets, based on the detection of data packet collisions, are segregated. Further, the segregation of these functionalities reduces the latency observed in the communication of the data packets from these network devices to other devices to which these network devices may be connected through a network.