Cloud Native SDN Architecture Multi-Cluster Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current software-defined networking (SDN) architectures face challenges in cloud-native adoption due to complexity in lifecycle management, scale limitations in configuration modules, and the lack of command-line interface (CLI)-based interfaces, which hinder efficient management and scalability in cloud data centers.

Innovation Solution

A cloud-native SDN architecture is deployed using multiple clusters with a centralized configuration and control plane, incorporating data plane elements in compute nodes and network devices, and a network controller that supports custom resources and microservices architecture, enabling scalable and elastic network management through a unified intent model and aggregated API layer.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional SDN architecture is used, then network control functionality is provided, but lifecycle management complexity increases and scalability is limited

Engineering Contradiction:
ImprovescalabilityVSAvoidlifecycle management complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The SDN controller is segmented into multiple independent controller instances distributed across different clusters. Each controller manages a specific cluster, allowing independent scaling and management. This segmentation reduces lifecycle management complexity by enabling granular control and parallel operations across clusters.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The architecture introduces a multi-dimensional control plane by adding cluster-level abstraction above individual controller instances. This dimensional addition enables centralized policies to be applied across multiple clusters while maintaining individual cluster autonomy, resolving the contradiction between scalability and management complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Adaptability or versatility

If SDN architecture supports multiple clusters, then scalability improves, but configuration management becomes more complex

Engineering Contradiction:
Improvemulti-cluster supportVSAvoidconfiguration management complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

A unified configuration management system is implemented that provides universal functionality across all clusters. The system uses standardized configuration schemas and policies that can be applied consistently across multiple clusters, reducing configuration management complexity while maintaining multi-cluster support.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

A centralized configuration service acts as an intermediary between administrators and distributed controllers. This mediator abstracts the complexity of multi-cluster configuration by providing a unified interface for policy definition and distribution, automatically translating high-level configurations into cluster-specific settings.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Productivity

If centralized control plane is implemented, then network management efficiency improves, but single point of failure risk increases

Engineering Contradiction:
Improvenetwork management efficiencyVSAvoidfault tolerance
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The centralized control plane is segmented into multiple distributed controller instances across different clusters. Each instance maintains independent operation capability, allowing the system to achieve centralized management efficiency while distributing failure points. If one controller fails, others continue to operate independently.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The architecture dynamically adjusts the degree of centralization based on operational parameters. During normal operation, controllers coordinate closely for efficient centralized management. During failure conditions, controllers transition to more autonomous operation, changing the system's organizational parameter to maintain both efficiency and reliability.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4160409A1Cloud native software-defined network architecture for multiple clusters
Publication Date: 2023.04.05 JUNIPER NETWORKS INC
  • EP4160409A1 patent drawingFigure 1
  • EP4160409A1 patent drawingFigure 2
  • EP4160409A1 patent drawingFigure 3

AI summary

In an example, a network controller for an SDN architecture system includes processing circuitry of a central cluster of first one or more compute nodes; a configuration node configured for execution by the processing circuitry; a control node configured for execution by the processing circuitry. The configuration node includes a custom API server to process requests for operations on custom resources for SDN architecture configuration. Each of the custom resources for SDN architecture configuration corresponds to a type of configuration object in the SDN architecture system. The control node, in response to detecting an event on an instance of a first custom resource of the custom resources, obtains configuration data for the instance of the first custom resource and configures a corresponding instance of a configuration object in a workload cluster of second one or more compute nodes. The first and second compute nodes may be distinct.