Packet Delivery Rule Prioritization in Network Switches
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current packet delivery methods in network switches, especially in 5G mobile communication networks, face delays due to the need for software processing when packet delivering rules are not found in ternary content-addressable memory (TCAM), which is unfavorable for applications requiring low network latency.
Innovation Solution
Designating priorities to packet delivering rules allows for strategic loading from the software layer to the hardware layer, with a mechanism to replace rules in the hardware layer when full, ensuring efficient storage and reducing delay by prioritizing higher-priority rules for direct hardware processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If packet delivering rules are stored only in hardware layer (TCAM), then packet delivery speed is improved, but the number of storable rules is limited and lower-priority rules cannot be accommodated
Solution Approach 1:
The patent divides packet delivering rules into two segments: high-priority rules stored in hardware layer (TCAM) for fast processing, and low-priority rules stored in software layer for later loading. This segmentation allows the system to maintain fast hardware processing while accommodating a larger total number of rules in software, resolving the contradiction between speed and quantity.
Solution Approach 2:
Different storage locations (hardware vs software) are assigned different qualities based on priority. Hardware layer stores only high-priority rules requiring fast access, while software layer stores low-priority rules that can tolerate slower access. This local quality differentiation optimizes both speed for critical rules and quantity for overall rule capacity.
2Loss of time
If packet delivering rules are loaded from software layer to hardware layer, then packet delivery delay is reduced, but hardware storage space is limited
Solution Approach 1:
The system performs preliminary classification of rules by priority before loading them to hardware. High-priority rules are pre-loaded to hardware layer in advance, ensuring they are ready for fast processing when needed, while low-priority rules remain in software layer. This preliminary action optimizes hardware usage and reduces packet delivery delay for critical rules.
Solution Approach 2:
The patent implements dynamic rule loading and replacement mechanisms. When hardware storage is full, the system dynamically replaces low-priority rules with newly arrived high-priority rules. This dynamic adjustment allows the system to adapt to changing traffic patterns and maintain optimal performance despite limited hardware storage space.
3Adaptability or versatility
If all packet delivering rules are processed by software layer, then rule flexibility is maintained, but network delay increases
Solution Approach 1:
The patent creates a universal packet delivery system that can handle both high-priority and low-priority rules through a unified architecture. The hardware layer provides fast path for high-priority rules, while the software layer handles low-priority rules and can dynamically load high-priority rules to hardware. This multi-functional design maintains rule flexibility while reducing delay for time-sensitive traffic.
Solution Approach 2:
The controller acts as an intermediary between software and hardware layers, managing rule loading, replacement, and priority assignments. This intermediary coordinates the interaction between the flexible software layer and the fast hardware layer, ensuring that rules are optimally distributed to balance flexibility and speed requirements.
Data Source
AI summary
A packet delivering method for a packet delivering apparatus includes: recording respective priorities of multiple packet delivering rules, which includes multiple first packet delivering rules and multiple second packet delivering rules respectively stored in a software layer and a main storage space of a hardware layer of the packet delivering apparatus; selecting a designated packet delivering rule from the first packet delivering rules; determining whether the main storage space is full; searching for at least one candidate delivering rule that meets a replacement condition from the second packet delivering rules according to the priorities of the designated data of the designated packet delivering rule and the second packet delivering rules when the main storage space is full; and replacing one of the at least one candidate packet delivering rule by the designated packet delivering rule when the at least one candidate packet delivering rule exists.