Adaptive MAC Timeout in EVPN Using Machine Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing Ethernet virtual private network (EVPN) technologies face challenges in efficiently managing media access control (MAC) address timeouts, leading to unnecessary MAC addresses remaining in forwarding tables for longer than needed, causing ARP flooding and inefficient resource utilization.

Innovation Solution

A device monitors EVPN traffic to generate a machine learning-based traffic model for MAC addresses, determining customized timeouts based on predicted traffic patterns, and causes the MAC addresses to be timed out from forwarding tables, reducing unicast and ARP flooding.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a fixed timeout value is used for all MAC addresses in EVPN forwarding tables, then the system is simple to manage, but resources are wasted by keeping inactive MAC addresses in the table for unnecessary durations

Engineering Contradiction:

ImproveMAC timeout management complexityVSAvoidresource utilization efficiency

Core Design Contradiction:

Device complexityVSLoss of energy

Solution Approach 1:

The patent applies dynamics by transitioning from a fixed timeout value to a dynamic timeout mechanism that adapts to individual MAC address traffic patterns. The system continuously monitors traffic behavior and adjusts timeout values accordingly, allowing the timeout to vary based on observed communication activity rather than using a static predetermined value for all MAC addresses.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent implements parameter changes by modifying the timeout parameter based on traffic pattern analysis. The system changes the timeout value dynamically according to the observed traffic characteristics of each MAC address, such as communication frequency and duration patterns, thereby optimizing resource utilization while maintaining simplicity in management.

Inventive Principle:

Principle #35Parameter changes

2Reliability

If MAC addresses are kept in forwarding tables for extended periods to ensure connectivity, then connectivity reliability is improved, but ARP flooding increases and resource utilization deteriorates

Engineering Contradiction:

Improveconnectivity reliabilityVSAvoidARP flooding

Core Design Contradiction:

ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent applies feedback by implementing a monitoring mechanism that observes actual traffic patterns of MAC addresses and uses this information to adjust timeout values. The system feedback-loop continuously tracks communication activity and modifies timeout parameters accordingly, ensuring that MAC addresses remain in the table only as long as traffic patterns indicate connectivity is needed, thereby reducing ARP flooding while maintaining reliability.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent implements preliminary action by proactively adjusting timeout values based on predicted traffic patterns rather than reacting to actual traffic occurrence. The system analyzes historical traffic data to predict future communication needs and sets appropriate timeout values in advance, preventing premature removal of MAC addresses while avoiding unnecessary retention, thus reducing ARP flooding proactively.

Inventive Principle:

Principle #10Preliminary action

3Reliability

If timeout values are extended for frequently communicating MAC addresses, then connectivity is enhanced, but the system complexity increases due to need for traffic monitoring and analysis

Engineering Contradiction:

Improveconnectivity for frequently communicating endpointsVSAvoidtraffic monitoring and analysis complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies self-service by enabling the system to automatically monitor, analyze, and adjust timeout values without external intervention. The system performs self-optimization by internally tracking traffic patterns and autonomously modifying timeout parameters based on observed behavior, thereby enhancing connectivity for frequently communicating endpoints while avoiding the need for complex external management systems.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual or rigid mechanical timeout management with an intelligent system that uses traffic pattern analysis and machine learning algorithms. Instead of fixed mechanical rules, the system substitutes automated adaptive mechanisms that learn from traffic behavior and dynamically adjust timeouts, reducing the need for complex manual configuration while improving connectivity for active endpoints.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10313153B2Adaptive MAC grouping and timeout in EVPN environments using machine learning

Publication Date: 2019.06.04 CISCO TECHNOLOGY INC

AI summary

In one embodiment, a device in a network monitors Ethernet virtual private network (EVPN) traffic in the network for a plurality of media access control (MAC) addresses. The device generates a machine learning-based traffic model for the MAC addresses based on the monitored EVPN traffic. The device determines a timeout for a particular one of the MAC addresses based on traffic predicted by the machine learning-based traffic model for the particular MAC address. The device causes the particular MAC address to be timed out from one or more forwarding tables in the network based on the determined timeout.