Polymorphic Network Switching Resource Allocation via Machine Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Network programmers face challenges in effectively utilizing switching resources in polymorphic networks due to the complexity of interactions between applications and hardware, leading to inefficient resource allocation and potential errors.

Innovation Solution

A method utilizing machine learning for optimizing switching resource allocation, involving manual pre-configuration, offline learning, and online reasoning to dynamically adjust network modality packet forwarding rules, selecting between ASIC, FPGA, and PPK based on performance indices and classifier models to guide optimal resource allocation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If manual configuration of switching resources is performed, then network programmers can specify resource allocation, but the complexity of interaction between applications and hardware makes it difficult to decide when and how to effectively use switching resources

Engineering Contradiction:

Improveease of switching resource allocationVSAvoidcomplexity of application-hardware interaction

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The system enables self-service by allowing the network element to automatically determine and allocate switching resources based on packet characteristics and pre-configured policies, eliminating the need for programmers to manually decide resource allocation for each application-hardware interaction scenario

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent introduces an intermediary layer (the switching resource allocation mechanism) that sits between the application layer and hardware layer, automatically making allocation decisions based on packet classification and pre-configured policies, thus shielding programmers from the complexity of direct application-hardware interaction

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If switching resources are allocated manually, then resource usage can be controlled, but errors and costs in switching resource allocation may occur

Engineering Contradiction:

Improveaccuracy of switching resource allocationVSAvoidcomplexity of resource allocation management

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system performs self-service by automatically classifying packets and allocating switching resources based on pre-configured policies and real-time packet characteristics, eliminating human error in allocation decisions while maintaining high reliability

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system implements feedback mechanisms where packet classification results and resource allocation outcomes are continuously monitored and used to adjust allocation decisions, ensuring accurate and reliable resource distribution without manual intervention

Inventive Principle:

Principle #23Feedback

3Adaptability or versatility

If software switches are used for all network modalities, then flexibility is achieved, but processing speed decreases

Engineering Contradiction:

Improveflexibility of network modality processingVSAvoidpacket forwarding speed

Core Design Contradiction:

Adaptability or versatilityVSSpeed

Solution Approach 1:

The system dynamically selects between software and hardware switching based on packet characteristics and current system state, allowing the network element to adapt its processing mode in real-time to achieve both flexibility and high speed for different network modalities

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent applies local quality by assigning different processing modes to different packet types or network modalities, where critical time-sensitive traffic uses hardware switching for speed while other traffic uses software switching for flexibility, optimizing overall system performance

Inventive Principle:

Principle #3Local quality

4Speed

If hardware switches are used for all packet processing, then processing speed increases, but flexibility to handle new network modalities decreases

Engineering Contradiction:

Improvepacket forwarding speedVSAvoidability to handle new network modalities

Core Design Contradiction:

SpeedVSAdaptability or versatility

Solution Approach 1:

The system dynamically determines the appropriate switching mode for each packet or flow based on classification results and current capabilities, allowing hardware acceleration for known modalities while falling back to software processing for new or unsupported modalities, thus maintaining both speed and flexibility

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS12056533B2Method, apparatus and medium for optimizing allocation of switching resources in polymorphic network

Publication Date: 2024.08.06 ZHEJIANG LAB

AI summary

A method, an apparatus and a medium for optimizing allocation of switching resources in the polymorphic network. The method selects the ASIC switching chip, FPGA and PPK software switching on the polymorphic network element based on machine learning, and specifically comprises the following steps: manually pre-configuring, formulating basic rules for polymorphic software and hardware co-processing; offline learning, designing training configuration in the offline learning stage to capture different switching resource usage variables, running experiments to generate the original data of a training classifier, and using the generated performance indices to train the model offline; and online reasoning, obtaining switching resource allocation advises, and updating modality codes according to the switching resource allocation advises.