Network Hardware Synchronization via FTOS for Seamless Failover

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Efficiently synchronizing hardware units in distinct processing domains of a network device is challenging, leading to increased complexity, cost, and performance degradation due to the need for additional hardware and software resources, especially when handling multiple threads and threads synchronization.

Innovation Solution

Implementing a fault-tolerant operating system (FTOS) on the network device to offload the synchronization process, using a high-speed transport protocol (HSTP) for message exchange between active and standby hardware units, with a message descriptor data structure (MDDS) for message management, ensuring transparent synchronization without awareness from the hardware units.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If hardware units synchronize directly with each other in distinct processing domains, then synchronization reliability is improved, but device complexity and cost increase due to additional hardware and software resources

Engineering Contradiction:
Improvesynchronization reliabilityVSAvoiddevice complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent introduces a shared memory region as an intermediary between hardware units in distinct processing domains. Instead of direct synchronization requiring complex inter-communication mechanisms, units write to and read from the shared memory through a mediator interface, simplifying the synchronization architecture while maintaining reliability.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent uses memory copying mechanisms where data is copied between processing domains through the shared memory interface. This allows synchronization without requiring complex direct communication channels, reducing device complexity while ensuring data consistency across domains.

Inventive Principle:
Principle #26Copying

2Measurement precision

If hardware units synchronize directly with each other, then synchronization accuracy is improved, but resource utilization increases leading to performance degradation

Engineering Contradiction:
Improvesynchronization accuracyVSAvoidperformance
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent enables hardware units to perform self-synchronization by autonomously reading from and writing to the shared memory region without requiring continuous intervention from synchronization controllers. This reduces the resource overhead associated with active synchronization management while maintaining accurate synchronization states.

Inventive Principle:
Principle #25Self-service

3Reliability

If additional hardware resources are allocated for synchronization, then synchronization reliability is improved, but cost increases

Engineering Contradiction:
Improvesynchronization reliabilityVSAvoidhardware resources
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The shared memory region serves multiple functions: it acts as both a data storage area and a synchronization coordination mechanism. This multi-functionality eliminates the need for dedicated synchronization hardware resources, reducing overall hardware quantity while maintaining reliability through the dual-purpose memory structure.

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

Data Source

PatentUS12634243B2Efficient high availability within a network device
Publication Date: 2026.05.19 HEWLETT PACKARD ENTERPRISE DEV LP
  • US12634243B2 patent drawing
  • US12634243B2 patent drawing
  • US12634243B2 patent drawing

AI summary

A network device is provided. During operation, the network device receives, from a first hardware unit of the network device, a message indicating an operation to be performed at the network device. Here, the first hardware unit includes processing resources distinct from the processing resources of the network device. The network device performs the operation at the network device based on information in the message. The network device sends an output of the operation to a second hardware unit of the network device that operates as a standby for the first hardware unit for facilitating high availability. The network device then updates the second hardware unit with the output of the operation to synchronize between the first and second hardware units via the processing resources of the network device. Subsequently, the network device provides, to the first hardware unit, an acknowledgment of the completion of the operation.