FCoE Data Forwarder State Management for High Availability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current High Availability (HA) solutions for Fiber Channel over Ethernet (FCoE) networks face issues due to the linear increase in cost with the number of switches in an HA cluster, primarily because they require synchronization of connection state between primary and secondary FCoE Forwarders, which is inefficient and costly.

Innovation Solution

The solution involves maintaining connection state information in FCoE Data Forwarders (FDFs), allowing any adjacent FCoE Forwarder (FCF) to replace a failed controlling FCF without the need for state synchronization, thereby reducing the number of redundant FCFs required and minimizing the cost structure to (1/N-1) where N is the number of switches in the HA cluster.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If connection state is maintained in primary FCF and synchronized to secondary FCF, then high availability is achieved, but cost increases linearly with number of switches in HA cluster

Engineering Contradiction:
Improvehigh availabilityVSAvoidnumber of redundant FCFs
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent extracts the connection state information from the FCF and places it in the FDF. This allows any adjacent FCF to immediately assume control of a failed FCF's FDFs without requiring state synchronization, eliminating the need for expensive primary/secondary FCF pairs and reducing the number of redundant FCFs from linear to constant (1/N-1).

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the control architecture by separating FDF control from FCF control. Each FDF is independently controlled by its adjacent FCF, eliminating the need for centralized state synchronization across the HA cluster. This segmentation allows independent failover of each FDF-FCF pair without affecting others.

Inventive Principle:
Principle #1Segmentation

2Reliability

If connection state synchronization is implemented between primary and secondary FCF, then failover capability is provided, but operational complexity and cost increase

Engineering Contradiction:
Improvefailover capabilityVSAvoidstate synchronization mechanism
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent removes the state synchronization mechanism entirely by extracting connection state information from the FCF and storing it in the FDF. This eliminates the complex synchronization protocol between primary and secondary FCFs while maintaining failover capability through direct FDF control by adjacent FCFs.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Each FDF maintains its own connection state information locally, making it self-sufficient during failover. When an FCF fails, the adjacent FCF can immediately take over the FDF's connections using the locally stored state information without requiring external synchronization or complex coordination mechanisms.

Inventive Principle:
Principle #25Self-service

3Ease of operation

If per connection state is maintained in primary FCF, then connection management is centralized, but scalability is limited by synchronization overhead

Engineering Contradiction:
Improveconnection managementVSAvoidscalability
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent segments connection management by allowing each FDF to be independently controlled by its adjacent FCF. This eliminates the centralized state synchronization bottleneck, enabling the network to scale linearly with the number of FDFs without proportional increases in synchronization overhead or operational complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Connection state information is pre-stored in each FDF, eliminating the need for real-time synchronization during failover events. This preliminary preparation of state information in the FDF enables immediate failover action without coordination overhead, significantly improving scalability.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS10367681B2Maintenance of data forwarder connection state information
Publication Date: 2019.07.30 HEWLETT PACKARD ENTERPRISE DEV LP
  • US10367681B2 patent drawing
  • US10367681B2 patent drawing
  • US10367681B2 patent drawing

AI summary

An example of high availability for FC or FCoE can include maintaining connection state information in a Fiber Channel over Ethernet (FCoE) Data Forwarder (FDF) or Fiber Channel Data Forwarder (FcDF). A failure of a first FCoE Forwarder (FCF) associated with the FDF can be detected. A second FCF can check a configuration file of the second FCF to see if the second FCF is responsible for the FDF. The second FCF can request the virtual domain ID, of the failed FCF, from a principal switch. The second FCF can be used as a service node for the FDF.