Multi-Memory Packet Classification System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network systems face inefficiencies in packet classification and processing, which hinders advanced network services such as routing, access control, and intrusion detection, due to limitations in handling multiple protocol layers and securing network communications.
Innovation Solution
A system with a data navigation unit and rule processing units that navigate and process rules stored in memory to classify packets based on their headers, using a tree structure and rule match engines to determine matches, allowing for efficient processing across various network protocol layers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If packet classification is performed using traditional single-memory structures, then device complexity is reduced, but processing speed and classification accuracy deteriorate
Solution Approach 1:
The patent divides the classification memory into multiple segments (first memory and second memory), each storing different portions of the classification rules. This segmentation allows parallel access to different rule sets, improving processing speed while distributing the complexity across separate memory structures rather than requiring a single complex memory system.
Solution Approach 2:
The patent introduces a hierarchical dimension to the memory structure by organizing rules across multiple memory levels (first memory for frequently accessed rules, second memory for less frequently accessed rules). This dimensional organization enables faster access to critical classification rules while maintaining comprehensive rule coverage, thus improving speed without proportionally increasing overall complexity.
2Reliability
If multiple protocol layers are processed simultaneously for comprehensive security, then security coverage is improved, but processing time increases
Solution Approach 1:
The patent pre-organizes classification rules in a hierarchical structure before packet arrival, with general classification rules in the first memory and specific protocol-layer rules in the second memory. This preliminary organization allows the system to quickly determine which protocol layers require processing and retrieve the appropriate rules in advance, maintaining comprehensive security coverage while reducing actual processing time.
Solution Approach 2:
The patent implements selective processing where not all protocol layers are processed with equal depth for every packet. Instead, the system performs partial processing based on packet characteristics, retrieving only the necessary rules from the hierarchical memory structure. This approach maintains adequate security coverage while avoiding the time cost of exhaustive multi-layer processing for every packet.
3Adaptability or versatility
If comprehensive packet classification rules are stored in a single memory, then rule availability is improved, but access time increases
Solution Approach 1:
The patent segments the comprehensive classification rules into two separate memory structures: the first memory stores general and frequently accessed rules, while the second memory stores specific and less frequently accessed rules. This segmentation maintains complete rule coverage for versatile classification while reducing access time for the most common classification operations by keeping them in the faster first memory.
Solution Approach 2:
The patent assigns different quality characteristics to different memory portions, with the first memory optimized for fast access to critical rules and the second memory optimized for storing comprehensive but less frequently accessed rules. This local quality differentiation ensures that the most important classification operations are performed quickly while maintaining overall rule comprehensiveness, thus achieving both versatility and speed.
Data Source
AI summary
In one embodiment, a system includes a data navigation unit configured to navigate through a data structure stored in a first memory to a first representation of at least one rule. The system further includes at least one rule processing unit configured to a) receive the at least one rule based on the first representation of the at least one rule from a second memory to one of the rule processing unit, and b) processing a key using the at least one rule.