Ethernet Packet Classification for Dynamic Cut-Through Forwarding

Solution Overview

Problem

Existing industrial control systems face challenges in adapting to multiple communication protocols and achieving real-time data packet classification with low latency and high throughput, particularly in industrial Ethernet environments where protocols like EtherCAT and Profinet require efficient parsing and forwarding of data packets.

Innovation Solution

The proposed solution involves a network communications interface, memory, and processing units that classify data packets based on factors like rate and time of receipt, allowing for the selection of either cut-through or store-and-forward operations, and utilizing hardware accelerators and programmable real-time units (PRUs) to perform operations like checksum calculation and data transfer with low latency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If store-and-forward operation is used for data packet processing, then data integrity is improved, but processing latency increases

Engineering Contradiction:

Improvedata integrityVSAvoidprocessing latency

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system dynamically selects between store-and-forward and cut-through operations based on real-time classification of data packets. The classification unit evaluates packet characteristics and determines the appropriate forwarding mode, allowing the system to adapt its processing behavior to balance reliability and latency requirements for different traffic types.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the processing parameter (forwarding mode) based on packet classification. By adjusting the operational mode from static to dynamic, the system can optimize for either data integrity or processing speed depending on the specific packet requirements and network conditions.

Inventive Principle:

Principle #35Parameter changes

2Loss of time

If cut-through operation is used for data packet processing, then processing latency is reduced, but data integrity may be compromised

Engineering Contradiction:

Improveprocessing latencyVSAvoiddata integrity

Core Design Contradiction:

Loss of timeVSReliability

Solution Approach 1:

The system dynamically selects between store-and-forward and cut-through operations based on real-time classification of data packets. The classification unit evaluates packet characteristics and determines the appropriate forwarding mode, allowing the system to adapt its processing behavior to balance reliability and latency requirements for different traffic types.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the processing parameter (forwarding mode) based on packet classification. By adjusting the operational mode from static to dynamic, the system can optimize for either data integrity or processing speed depending on the specific packet requirements and network conditions.

Inventive Principle:

Principle #35Parameter changes

3Adaptability or versatility

If multiple communication protocols are supported, then adaptability is improved, but device complexity increases

Engineering Contradiction:

Improvemulti-protocol supportVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The classification unit is designed to handle multiple communication protocols (EtherCAT, Profinet, Ethernet IP, Modbus TCP) through a unified architecture. This universal approach allows the system to support diverse protocols without requiring separate dedicated processing paths for each protocol, thereby managing complexity while maintaining versatility.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system segments the protocol handling into distinct classification stages within the classification unit. By dividing the complex protocol processing into manageable segments (parsing, classification, forwarding decision), the system can support multiple protocols without overwhelming complexity in any single processing component.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11343205B2Real-time, time aware, dynamic, context aware and reconfigurable ethernet packet classification

Publication Date: 2022.05.24 TEXAS INSTRUMENTS INC

AI summary

An ultra-high speed electronic communications device includes: a network communications interface; a memory; and one or more processing units, communicatively coupled to the memory and the network communications interface, wherein the memory stores instructions configured to cause the one or more processing units to: receive a data packet using the network communications interface; determine a classification of the data packet based, at least in part, on a plurality of factors, wherein the plurality of factors comprises a rate at which the data packet was received and a time at which the data packet was received; select, based at least in part, on the classification, an operation from a plurality of operations, wherein the plurality of operations comprises a cut-through operation and a store-and-forward operation; and perform the selected operation.