Cell Tree Forest Packet Classification Engine
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Static packet classification in routers faces challenges with multi-dimensional extended search, requiring efficient data structures that balance search speed and memory usage, especially when handling a large number of concurrent traffic flows and frequent updates.
Innovation Solution
The implementation of a Cell Tree Forest (CTF) framework, which enhances Patricia trees to support storage of arbitrary cells and employs techniques like level compression and redundant tree elimination to improve search speed and reduce memory requirements, enabling efficient multi-match classification and handling of wildcard rules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multi-dimensional extended search is used for packet classification, then classification accuracy is improved, but search speed deteriorates
Solution Approach 1:
The patent divides the multi-dimensional classification space into multiple one-dimensional sorted lists, each representing a different dimension (e.g., source IP, destination IP, protocol). This segmentation allows the system to perform efficient one-dimensional binary searches instead of complex multi-dimensional searches, resolving the contradiction between classification accuracy and search speed.
Solution Approach 2:
The patent transforms the multi-dimensional classification problem into multiple one-dimensional problems by creating separate sorted lists for each dimension. This dimensionality reduction enables the use of efficient binary search algorithms while maintaining the ability to perform accurate multi-field packet classification.
2Adaptability or versatility
If data structure size is increased to handle more traffic flows, then classification completeness is improved, but memory usage increases
Solution Approach 1:
The patent segments the large classification data structure into multiple smaller one-dimensional sorted lists, each optimized for a specific dimension. This segmentation reduces the memory footprint of individual data structures while collectively maintaining the ability to classify all traffic flows, resolving the contradiction between classification completeness and memory usage.
Solution Approach 2:
The patent changes the structural parameters of the data structure from a single large multi-dimensional structure to multiple smaller one-dimensional structures. This parameter change optimizes memory utilization while preserving the classification capability for handling large numbers of concurrent traffic flows.
3Reliability
If frequent updates are performed to maintain classification accuracy, then classification reliability is improved, but processing overhead increases
Solution Approach 1:
The patent implements dynamic sorted lists that can be efficiently updated while maintaining their sorted property. The use of incremental update algorithms allows the data structures to adapt to changing traffic patterns and classification rules with minimal processing overhead, resolving the contradiction between classification reliability and processing complexity.
Data Source
AI summary
The disclosure relates to a static packet classification engine for data packet classification in a data communications network. There is disclosed a static packet classification engine configured to handle rules comprising wildcards and interval matching which is cell tree forest based configured to search clusters to produce results and comprise a priority encoder configured to process the results in terms of matches.