ML-Infused Network Topology Deployment for Hybrid Cloud

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Hybrid cloud computing environments are complex and costly to deploy and maintain due to functional disparities between different cloud environments and dynamic alterations, requiring complex filters and special-purpose virtual network elements, which complicates system management and scalability.

Innovation Solution

A topology deployment system that receives operational data from network topologies across multiple workload resource domains, uses machine-learning models to evaluate performance and provide optimization recommendations, and updates the topologies based on this data, along with resource inventories and constraints, to optimize and modify the network configurations dynamically.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If deployment models are used to improve consistency and uniform distribution of functions across cloud environments, then functional consistency is improved, but device complexity and cost increase due to requiring complex filters and special-purpose virtual network elements

Engineering Contradiction:

Improvefunctional consistencyVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent uses software-defined networking (SDN) controllers that create virtual copies of network functions and policies across multiple cloud environments. These software-based virtual network elements replicate the behavior and configuration of physical network devices, enabling consistent network functionality across hybrid cloud infrastructures without requiring specialized hardware in each environment.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The SDN controller architecture provides universal network management capabilities that can be applied across diverse cloud environments (public, private, hybrid). A single SDN controller can manage multiple network topologies, enforce consistent policies, and adapt to different underlying infrastructures, replacing the need for environment-specific specialized network elements.

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Reliability

If deployment models are used to ensure uniform distribution of functions, then functional consistency is improved, but maintenance cost increases due to complexity of deploying and maintaining across hybrid environments

Engineering Contradiction:

Improvefunctional consistencyVSAvoiddeployment and maintenance cost

Core Design Contradiction:

ReliabilityVSEase of manufacture

Solution Approach 1:

The SDN controller acts as an intermediary layer between the physical network infrastructure and the virtual network functions. This intermediary abstracts the complexity of hybrid cloud management, providing a unified interface for deployment and maintenance operations. Network administrators can manage consistent policies and configurations through the SDN controller without dealing with the underlying complexity of each cloud environment.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements automated feedback mechanisms where the SDN controller continuously monitors network performance and operational status across hybrid cloud environments. Based on this feedback, the system automatically adjusts and optimizes network configurations, reducing manual intervention and maintenance costs while ensuring consistent functional behavior across all environments.

Inventive Principle:

Principle #23Feedback

3Reliability

If special-purpose virtual network elements are used instead of traditional enterprise edge internetworking devices, then functional consistency across cloud environments is improved, but cost increases

Engineering Contradiction:

Improvefunctional consistencyVSAvoidcost

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The patent replaces traditional hardware-based enterprise edge internetworking devices with software-based virtual network elements managed by SDN controllers. This substitution transitions from mechanical/physical network infrastructure to software-defined networking, enabling consistent network functionality across cloud environments while reducing hardware costs and leveraging existing cloud infrastructure.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The SDN-based virtual network elements provide universal functionality that can operate across diverse cloud infrastructures without requiring specialized hardware. A single software-based virtual network element can serve multiple cloud environments, reducing the total quantity of network devices needed and lowering overall system cost while maintaining functional consistency.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS11424989B2Machine-learning infused network topology generation and deployment

Publication Date: 2022.08.23 CISCO TECHNOLOGY INC

AI summary

Techniques are described herein for deploying, monitoring, and modifying network topologies comprising various computing and network nodes deployed across multiple workload resource domains. A deployment system may receive operational data from a network topology deployed across multiple workload resource domains, such as public or private cloud computing environments, on-premise data centers, and the like. The operational data may be provided to a trained machine-learning model, and output from the trained model may be used, along with constraint inputs and resource inventories of the workload resource domains, to determine updated topology models which may be deployed within the workload resource domains.