SDN Forwarding Table Performance Control via Parameter Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional software defined networking (SDN) lacks a mechanism to control the performance of forwarding tables, leading to resource wastage and potential network failures due to unmanaged performance requirements and resource overutilization by higher priority forwarding tables.

Innovation Solution

A method where a control plane determines and distributes performance requirements to a forwarding plane, enabling the generation of trie trees and hash tables with optimized depth, load factor, and collision resolution algorithms to satisfy specific performance criteria, ensuring efficient resource usage and network performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If forwarding tables are implemented without performance control mechanisms, then device complexity is reduced and ease of operation is improved, but resource wastage increases and reliability deteriorates

Engineering Contradiction:

Improvenetwork performance reliabilityVSAvoidresource wastage

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The patent applies parameter changes by dynamically adjusting forwarding table parameters (such as table size, depth, and allocation) based on performance requirements. The control plane determines optimal parameters for each forwarding table and distributes them to the forwarding plane, ensuring that resources are allocated efficiently according to actual performance needs rather than using fixed or excessive allocations.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If higher priority forwarding tables over utilize hardware resources, then their performance is improved, but lower priority forwarding tables suffer resource starvation and their performance deteriorates

Engineering Contradiction:

Improveforwarding table performanceVSAvoidservice level agreement compliance

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent implements dynamics by enabling the control plane to dynamically determine and adjust performance requirements for different forwarding tables based on service level agreements and network conditions. This allows resource allocation to be flexible and adaptive, ensuring that high-priority tables receive necessary resources while low-priority tables are guaranteed minimum resources, preventing complete resource starvation.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system employs feedback mechanisms where the control plane monitors forwarding table performance and resource utilization, then adjusts performance requirements and resource allocations accordingly. This closed-loop control ensures that service level agreements are maintained and that resource distribution remains balanced across multiple forwarding tables.

Inventive Principle:

Principle #23Feedback

3Manufacturing precision

If trie tree depth is increased to improve search accuracy, then manufacturing precision is improved, but the number of steps increases and productivity deteriorates

Engineering Contradiction:

Improveforwarding table search accuracyVSAvoidpacket forwarding speed

Core Design Contradiction:

Manufacturing precisionVSProductivity

Solution Approach 1:

The patent applies parameter changes by optimizing trie tree parameters (depth, stride, and allocation) based on performance requirements. The control plane determines the optimal depth and stride values that balance search accuracy with forwarding speed, distributing these optimized parameters to the forwarding plane for implementation.

Inventive Principle:

Principle #35Parameter changes

4Loss of substance

If hash table load factor is increased to reduce memory usage, then loss of substance is reduced, but hashing performance deteriorates due to collisions

Engineering Contradiction:

Improvememory resource usageVSAvoidhashing performance

Core Design Contradiction:

Loss of substanceVSProductivity

Solution Approach 1:

The patent applies parameter changes by optimizing hash table parameters (load factor, bucket size, and allocation) based on performance requirements. The control plane determines the optimal load factor that balances memory efficiency with hashing performance, taking into account the specific performance requirements of each forwarding table.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS9450866B2Forwarding table performance control in SDN

Publication Date: 2016.09.20 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

AI summary

Exemplary methods for controlling forwarding table performance include a first network device in a control plane determining a first performance requirement of a first forwarding table in a forwarding plane based on an overall performance requirement of a plurality forwarding tables in the forwarding plane. In one embodiment, in response to determining the first forwarding table in the forwarding plane can be generated to satisfy the first performance requirement, the methods include the first network device sending a first message that includes the first performance requirement to a second network device in the forwarding plane, causing the second network device to generate the first forwarding table that satisfies the first performance requirement. In one embodiment, the exemplary methods include the second network device generating the first forwarding table that satisfies the first performance requirement included in the first message.