Data Replication in Virtual Networking Systems

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In network virtualization using overlays, the incompatibility between different packet formats necessitates translation gateways, which often act as communication bottlenecks, impacting performance.

Innovation Solution

A method for data replication in a networking system with multiple computing nodes, where data updates are replicated across all nodes maintaining the data set, ensuring consistent updates and reducing reliance on translation gateways.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If translation gateways are used to translate between different packet formats, then compatibility between encapsulation types is achieved, but communication performance deteriorates due to bottlenecks

Engineering Contradiction:

Improvecompatibility between encapsulation typesVSAvoidcommunication performance

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent extracts the translation function from a centralized gateway and distributes it to individual computing nodes. Each node maintains local translation capabilities, eliminating the need for centralized translation gateways that create communication bottlenecks. This allows direct peer-to-peer translation between nodes with different encapsulation types.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The translation functionality is segmented from the gateway infrastructure and assigned to individual computing nodes. Each node independently handles translation for its own packets, dividing the translation workload across multiple nodes rather than concentrating it in a single gateway, thereby eliminating the bottleneck.

Inventive Principle:

Principle #1Segmentation

2Reliability

If data is maintained on multiple computing nodes, then system reliability and availability are improved, but data consistency becomes more difficult to maintain

Engineering Contradiction:

Improvesystem availabilityVSAvoiddata consistency

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The patent implements a feedback mechanism where computing nodes send notifications to a namespace service when data changes occur. The namespace service tracks data locations and notifies other nodes of changes, ensuring all nodes maintain consistent views of the data without requiring centralized control.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system pre-establishes data replication across multiple nodes before data changes occur. By maintaining copies of data on multiple nodes in advance and using the feedback mechanism to keep them synchronized, the system ensures data availability and consistency without requiring real-time coordination during data access.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10230795B2Data replication for a virtual networking system

Publication Date: 2019.03.12 KYNDRYL INC

AI summary

Embodiments of the invention provide a method for data replication in a networking system comprising multiple computing nodes. The method comprises maintaining a data set on at least two computing nodes of the system. The method further comprises receiving a data update request for the data set, wherein the data update request includes a data update for the data set. The data set on the at least two computing nodes is updated based on the data update request received.