Intermediary Ethernet Switch for FCoE Failure Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In Fibre Channel over Ethernet networks, ports connected to a failed switch cannot immediately detect the failure due to the lack of link down events, leading to inaccessible end ports and potential data loss.

Innovation Solution

Establishing an Ethernet connection among Fibre Channel over Ethernet switches with an intermediary Ethernet switch, allowing for direct connection avoidance between switches and end ports, and using protocols like multicast or unicast handshakes for port ID assignment and failure notifications, enabling rapid switch failover through Extended Link Service commands.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If ports are connected directly to Fibre Channel switches, then failure detection is immediate through link down events, but network accessibility is lost when switch fails

Engineering Contradiction:
Improvefailure detection accuracyVSAvoidnetwork accessibility
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

An intermediary Ethernet switch is introduced between end ports and FCoE switches. The end port connects to the Ethernet switch, which then connects to multiple FCoE switches. This intermediary allows the end port to access multiple FCoE switches through a single physical connection, enabling failover capability while maintaining continuous network accessibility even when one FCoE switch fails.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If ports are connected to intermediary Ethernet switch with redundant FCoE switches, then network accessibility is maintained during failures, but failure detection becomes delayed

Engineering Contradiction:
Improvenetwork accessibilityVSAvoidfailure detection accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

A feedback mechanism is implemented where FCoE switches send periodic heartbeat or keepalive signals to the Ethernet switch, and the Ethernet switch monitors these signals. When a FCoE switch fails to respond, the Ethernet switch detects the failure through the absence of expected feedback signals and can then redirect traffic to alternative FCoE switches, maintaining network accessibility while enabling timely failure detection.

Inventive Principle:
Principle #23Feedback

3Device complexity

If direct connection is established between FCoE switches and end ports, then configuration is simplified, but redundancy and failover capability are reduced

Engineering Contradiction:
Improvenetwork configuration complexityVSAvoidswitch failover capability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The Ethernet switch is designed to perform multiple functions: it acts as a standard Ethernet switch for data forwarding, a failure detection monitor through heartbeat signal reception, and a failover coordinator that redirects traffic when FCoE switches fail. This multi-functional design enables redundancy and automatic failover without significantly increasing configuration complexity for the end user.

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

Data Source

PatentEP2361467B1Redundant intermediary switch solution for detecting and managing fibre channel over ethernet (FCOE) switch failures
Publication Date: 2012.09.05 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • EP2361467B1 patent drawingFigure 1
  • EP2361467B1 patent drawingFigure 2~3
  • EP2361467B1 patent drawingFigure 4

AI summary

A method for detecting a switch failure and managing switch failover in a Fibre Channel over Ethernet network. The plurality of Fibre Channel over Ethernet switches are connected to an intermediary Ethernet switch and to each other via an Ethernet connection. The intermediary Ethernet switch is different from the plurality of Fibre Channel over Ethernet switches. A plurality of end ports are also connected to the intermediary Ethernet switch. A failure is detected in a first switch in the plurality of Fibre Channel over Ethernet switches. The first switch is associated with a first end port in the plurality of end ports. A failure notification is sent to the first end port using the intermediary Ethernet switch. The failure notification might specify a second switch destination and an identification of the first switch. A fabric login is exchanged between the first end port and the second switch.