Open Network Switch Control Plane Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current network switches lack flexibility and resilience in managing data flows and software updates, as they rely on a centralized control plane that can be bottlenecked by software bugs or viruses, and require complex management procedures.

Innovation Solution

An integrated open network switch architecture with a separation of processing units, where a main processing unit runs a different OS than secondary and additional units, allowing for remote procedure calls and flexible application execution, enabling decentralized management and application migration between units without affecting the control plane.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a centralized control plane is used in network switches, then management and control functions can be consolidated, but the system becomes vulnerable to software bugs and viruses that can bottleneck the entire network

Engineering Contradiction:
Improvecontrol plane managementVSAvoidnetwork control reliability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent segments the control plane functions by introducing virtualization, where the control plane is divided into multiple virtual instances running on separate virtual machines. This allows the control plane to be distributed across multiple processing units, preventing a single point of failure and eliminating the bottleneck effect of centralized control while maintaining consolidated management capabilities.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a virtualization layer as an intermediary between the hardware control plane and the software applications. This virtualization layer isolates the control plane from direct software access, preventing viruses and bugs from compromising the underlying hardware while still allowing legitimate management functions to operate through the virtualized interface.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If custom applications are allowed on the control plane, then network functionality and adaptability improve, but system stability deteriorates due to potential software issues

Engineering Contradiction:
Improvenetwork application functionalityVSAvoidcontrol plane stability
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent segments the control plane into multiple isolated virtual machines, each capable of running custom applications. This segmentation allows diverse network applications to execute in isolated environments, improving adaptability while containing potential software issues within individual virtual instances rather than affecting the entire control plane.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements virtualization as a protective cushioning layer before custom applications can access the hardware control plane. This virtualization environment provides isolation and sandboxing, allowing custom applications to run with enhanced functionality while preventing them from directly impacting the stability of the underlying control plane infrastructure.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

3Adaptability or versatility

If software updates are applied to the control plane, then network management capabilities improve, but network operations are disrupted during the update process

Engineering Contradiction:
Improvecontrol plane software capabilitiesVSAvoidnetwork downtime during updates
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent segments the control plane into multiple virtual machines that can be independently updated. This allows software updates to be applied to individual virtual instances without requiring shutdown of the entire control plane, enabling continuous network operations during updates and eliminating downtime while still improving control plane capabilities.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic control plane virtualization where virtual machines can be created, updated, and migrated without stopping the control plane services. This dynamic approach allows the control plane to evolve and adopt new capabilities while maintaining continuous operation, preventing network disruptions during software updates.

Inventive Principle:
Principle #15Dynamics

4Reliability

If multiple processing units are used in the switch, then flexibility and resilience improve, but system complexity increases

Engineering Contradiction:
Improvesystem resilienceVSAvoidprocessing unit architecture
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent merges multiple processing units into a unified virtualized control plane environment. Instead of managing separate complex processing units independently, the virtualization layer consolidates their management into a single coherent system, reducing operational complexity while maintaining the resilience and flexibility benefits of having multiple processing units underlying the unified control plane.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS9367411B2System and method for an integrated open network switch
Publication Date: 2016.06.14 DELL SOFTWARE INC
  • US9367411B2 patent drawing
  • US9367411B2 patent drawing
  • US9367411B2 patent drawing

AI summary

A device includes a first processing unit and a second processing unit. The first processing unit is configured to execute a performance test on the device. The second processing unit is in communication with the first processing unit, and is configured to migrate an application from the second processing unit to the first processing unit. The second processing unit is further configured to detect a failure of the first processing unit, to migrate the application to a third processing unit in response to the failure of the first processing unit, and to assign a first plurality of ports to the third processing unit in response to the failure of the first processing unit.