Multi-Dimensional Packet Classification Using Address Space Compression

Solution Overview

Problem

Existing packet classification methods face challenges in efficiently handling larger IPv6 addresses, leading to increased board space, power consumption, and reduced lookup performance, particularly when compared to IPv4 classifiers.

Innovation Solution

The method employs address space compression to reduce the complexity of multi-dimensional classification by converting D-dimensional classification into single-field classifiers, reducing the number of bits required for classification and using TCAM-like memory for efficient matching.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multi-dimensional classification is performed on IPv6 addresses, then classification capability is improved, but board space and power consumption increase

Engineering Contradiction:

Improveclassification capabilityVSAvoidboard space

Core Design Contradiction:

Adaptability or versatilityVSArea of stationary object

Solution Approach 1:

The patent segments the multi-dimensional classification problem into multiple single-field classifications. Instead of performing one complex multi-dimensional classification on IPv6 addresses, the system divides it into several simpler single-field classification steps, each handling a specific dimension or field of the address. This segmentation reduces the computational complexity and resource requirements for each individual classification operation, thereby reducing board space and power consumption while maintaining overall classification capability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent transforms the classification problem from the address space dimension to a compressed dimension. By applying address space compression techniques, the system maps large IPv6 address spaces into smaller compressed representations, effectively changing the dimensionality of the classification problem. This allows the classifier to handle IPv6 addresses with reduced resource requirements, resolving the contradiction between classification capability and board space.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Adaptability or versatility

If multi-dimensional classification is performed on IPv6 addresses, then classification capability is improved, but power consumption increases

Engineering Contradiction:

Improveclassification capabilityVSAvoidpower consumption

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by stationary object

Solution Approach 1:

The patent segments the multi-dimensional classification problem into multiple single-field classifications. Instead of performing one complex multi-dimensional classification on IPv6 addresses, the system divides it into several simpler single-field classification steps, each handling a specific dimension or field of the address. This segmentation reduces the computational complexity and resource requirements for each individual classification operation, thereby reducing power consumption while maintaining overall classification capability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent transforms the classification problem from the address space dimension to a compressed dimension. By applying address space compression techniques, the system maps large IPv6 address spaces into smaller compressed representations, effectively changing the dimensionality of the classification problem. This allows the classifier to handle IPv6 addresses with reduced resource requirements, resolving the contradiction between classification capability and power consumption.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Adaptability or versatility

If multi-dimensional classification is performed on IPv6 addresses, then classification capability is improved, but lookup performance decreases

Engineering Contradiction:

Improveclassification capabilityVSAvoidlookup performance

Core Design Contradiction:

Adaptability or versatilityVSSpeed

Solution Approach 1:

The patent segments the multi-dimensional classification problem into multiple single-field classifications. Instead of performing one complex multi-dimensional classification on IPv6 addresses, the system divides it into several simpler single-field classification steps, each handling a specific dimension or field of the address. This segmentation reduces the computational complexity and resource requirements for each individual classification operation, thereby reducing power consumption while maintaining overall classification capability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent transforms the classification problem from the address space dimension to a compressed dimension. By applying address space compression techniques, the system maps large IPv6 address spaces into smaller compressed representations, effectively changing the dimensionality of the classification problem. This allows the classifier to handle IPv6 addresses with reduced resource requirements, resolving the contradiction between classification capability and power consumption.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS8477773B2Method, device and system for multi field classification in a data communications network

Publication Date: 2013.07.02 ALTERA CORP

AI summary

The present invention pertains to a method for performing specific data forwarding actions depending on the nature of data traffic comprising data packets, which method comprises the steps of: —receiving incoming data traffic of a specific nature, belonging to at least a specific class among a number of pre-defined classes, step 101—classifying the data traffic by determining the nature of the data traffic, step 102, provided by a process of inspecting values of one or more selected header field(s) of an address space of a data packet and selecting a first matching class from an ordered list of classes providing multi-dimensional classification, step 103.