Dynamic Prefix Management for White Box Router Memory Cost Reduction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

White box routers face high costs due to expensive memory requirements for forwarding tables, which store IP prefixes, and there is a need to dynamically manage and deploy subsets of prefixes efficiently to reduce resource allocation and prevent network errors.

Innovation Solution

A method and system that measure usage attributes of prefixes, such as frequency of use and volume of traffic, using machine learning to identify and store only the most used prefixes in a network device, allowing for dynamic management of prefix entries and reducing memory needs by using software-defined networking and a prefix usage analyzer.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If all IP prefixes are stored in the forwarding table memory, then routing completeness is improved, but memory cost and device complexity increase significantly

Engineering Contradiction:

Improverouting completenessVSAvoidmemory cost

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the forwarding table into two parts: a compact prefix table stored in fast memory (TCAM) containing only essential routing prefixes, and a longer prefix list stored in slower, cheaper memory. This segmentation allows the system to maintain routing completeness while significantly reducing the memory capacity required in expensive fast memory, thereby lowering overall device complexity and cost.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent extracts and identifies the most frequently used prefixes from the complete prefix set and stores them in the compact prefix table in fast memory. By taking out only the essential prefixes needed for common routing scenarios, the system maintains routing functionality for the majority of cases while avoiding the need to store all possible prefixes in expensive memory, thus resolving the contradiction between routing completeness and memory cost.

Inventive Principle:

Principle #2Taking out (Extraction)

2Quantity of substance

If larger fast memory is allocated to store more prefixes, then routing coverage is improved, but device cost increases significantly

Engineering Contradiction:

Improveprefix storage capacityVSAvoiddevice cost

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent implements a dynamic prefix management system that monitors traffic patterns and automatically adjusts which prefixes are stored in the compact prefix table. The system dynamically identifies frequently accessed prefixes and updates the prefix table accordingly, ensuring that the limited fast memory capacity is always utilized for the most valuable prefixes. This dynamic approach allows the system to maintain high routing coverage with minimal fast memory, avoiding the need to allocate large amounts of expensive memory statically.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of prefix selection criteria from static (storing all prefixes) to dynamic (storing only frequently used prefixes based on traffic analysis). By changing how prefixes are selected and stored, the system achieves effective routing coverage without requiring large fast memory capacity, thereby reducing device cost while maintaining functional requirements.

Inventive Principle:

Principle #35Parameter changes

3Reliability

If conservative resource allocation is used to prevent network errors, then network reliability is improved, but resource utilization efficiency deteriorates

Engineering Contradiction:

Improvenetwork reliabilityVSAvoidresource utilization efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent implements a feedback mechanism that continuously monitors network traffic patterns, prefix access frequencies, and routing performance. Based on this feedback, the system automatically adjusts the compact prefix table to include the most relevant prefixes, ensuring high routing accuracy and reliability. This feedback-driven approach eliminates the need for conservative over-allocation of memory resources, as the system adapts to actual network conditions and maintains reliability with optimized resource utilization.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs self-service by automatically monitoring its own performance, identifying frequently used prefixes, and updating the compact prefix table without external intervention. This self-adjusting mechanism ensures the system maintains high routing reliability while efficiently utilizing available memory resources, avoiding both conservative over-allocation and aggressive under-allocation of resources.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS11595308B2Closed loop prefix management and controller for whiteboxes

Publication Date: 2023.02.28 AT&T INTELLECTUAL PROPERTY I L P

AI summary

A system for closed loop prefix management for white boxes includes a network device, a route reflector coupled to the network device, a software defined network controller coupled to the route reflector and the network device, and a prefix usage analyzer in the software defined network controller. The prefix usage analyzer determines usage attributes of prefixes and identifies the prefixes with a predetermined usage attribute. The software defined network controller instructs a network controller in the network device to store the prefixes with the predetermined usage attributes in a table in the network device.