Fiber-Channel Fabric Principal Link Failover via Profile Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fibre channel fabrics face scalability limitations and increased system load due to the need for fabric reconfiguration when a principal link fails, leading to potential control frame drops, instability, and prolonged re-convergence times across all virtual SANs and switches.
Innovation Solution
A method and apparatus that detect failure modes of principal links, select and transmit profile information for redundant links, allowing seamless failover without propagating the failed link state throughout the network, thereby isolating the failed link and reducing processing load and convergence time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fabric reconfiguration is performed on all ports and switches when a principal link fails, then network redundancy is achieved, but system load increases and control frame drops occur
Solution Approach 1:
The patent segments the fabric reconfiguration process by identifying and isolating only the affected VSANs and ports when a principal link fails, rather than reconfiguring the entire fabric. This selective approach maintains network redundancy for affected segments while avoiding unnecessary processing load on the entire system.
Solution Approach 2:
The patent applies local quality by making reconfiguration actions localized to only those ports and VSANs directly impacted by the link failure. The failure impact is contained within specific fabric segments rather than propagating system-wide, reducing overall system load while maintaining reliability where needed.
2Reliability
If fabric reconfiguration is performed on all ports and switches when a principal link fails, then network redundancy is achieved, but control frame drops occur
Solution Approach 1:
The patent implements preliminary action by pre-establishing alternative paths and maintaining profile information for redundant links before failures occur. When a principal link fails, the system can immediately switch to pre-configured alternative paths without triggering widespread reconfiguration, thus preventing control frame drops while maintaining redundancy.
3Reliability
If fabric reconfiguration is performed on all ports and switches when a principal link fails, then network redundancy is achieved, but re-convergence times increase
Solution Approach 1:
The patent uses preliminary action by pre-configuring and storing profile information for alternative paths and redundant links before failures occur. This allows the system to perform rapid failover by simply activating pre-prepared configurations rather than computing new paths during failure events, significantly reducing re-convergence time while maintaining network redundancy.
4Adaptability or versatility
If scalability is increased to support more ports and virtual SANs on each switch, then network capacity is improved, but fabric stability decreases
Solution Approach 1:
The patent applies segmentation by dividing the fabric into independent VSAN segments, each with its own failure domain. This allows the fabric to scale to support more ports and VSANs while containing failure impacts within individual segments, preventing cascading effects that would otherwise destabilize the entire fabric as it grows in complexity.
Data Source
AI summary
In one embodiment, detecting a failure mode of a first communication link on a first port, selecting a second communication link on a second port associated with the first port, and transmitting a profile information associated with the second port are provided.