Logarithmic Packet Classifier Using Balanced Binary Trees
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network packet classification methods are inefficient in determining applicable policies for data packets due to the randomness in filter-ID searches, leading to suboptimal performance and increased search time.
Innovation Solution
Implementing a logarithmic time range-based multifield-correlation packet classifier using balanced binary trees and filter-IDs with bit mask descriptors to logically group and order policies, reducing search time by calculating intersection sets beforehand and using exact-match and range-based policy sets associated with each node.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If traditional packet classification methods are used with random filter-ID searches, then policy determination can be performed, but search time increases and performance deteriorates
Solution Approach 1:
The patent pre-calculates and stores intersection sets of policyIDs at each node of the balanced binary trees during the tree construction phase. This preliminary computation eliminates the need for runtime intersection calculations, reducing search time complexity from linear to logarithmic while maintaining classification accuracy
Solution Approach 2:
The patent divides the policy set into multiple balanced binary trees organized by filter element types (source address, destination address, source port, destination port). Each tree is independently structured with pre-computed intersection sets, allowing efficient segmented search that combines results logarithmically rather than requiring exhaustive linear search across all policies
2Reliability
If exhaustive search methods are used to determine policy matches, then all applicable policies can be found, but the complexity of the classification process increases
Solution Approach 1:
Intersection sets of policyIDs are pre-computed and stored at each node during tree construction. This eliminates complex runtime intersection calculations while ensuring all applicable policies are found, maintaining reliability without the computational overhead of exhaustive search
Solution Approach 2:
The patent uses dynamic balanced binary trees that automatically maintain optimal structure for search efficiency. The trees are rebuilt when policies are added or removed, adapting to changing conditions while preserving logarithmic search time complexity and complete policy match coverage
Data Source
AI summary
Classification of network data packets includes a determination sets of one or more filter-identifiers where each set is associated with a respective data-packet classifier field. A result-set of filter-identifiers may be derived based on an intersection of the filter-identifier sets.