Storage Processor Port Failover via Virtualization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing data storage systems require expensive and inconvenient host-based failover software to manage port failover in case of a single physical connection or storage processor failure, which can disrupt connectivity and data availability.

Innovation Solution

A method is introduced that uses a first storage processor to redirect requests from a second storage processor's virtual ports to its physical ports, allowing for seamless failover and failback processing without the need for host-based failover software, leveraging Fibre Channel N-Port ID Virtualization (NPIV) to dynamically move port identities and maintain continuous connectivity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If host-based failover software is used to manage port failover, then connectivity can be maintained during storage processor failure, but the system becomes expensive and inconvenient to deploy and maintain

Engineering Contradiction:

Improveconnectivity availabilityVSAvoidsoftware configuration complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The storage array performs automatic failover processing without requiring host-based software. The storage array detects failures and redirects requests internally using its own controllers, making the system self-sufficient for failover operations.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent introduces a coordination mechanism where one storage processor acts as an intermediary to redirect requests to another storage processor's physical ports. This intermediary role eliminates the need for external host-based failover software.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If multiple physical connections to storage processors are implemented, then continued connectivity is ensured during failure, but the cost and configuration complexity increase

Engineering Contradiction:

Improveconnectivity continuityVSAvoidconfiguration convenience

Core Design Contradiction:

ReliabilityVSEase of operation

Solution Approach 1:

The storage array automatically manages the configuration of multiple physical connections and failover paths without requiring manual host configuration. The system self-configures the redundant connections and failover behavior.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent combines multiple physical connections and failover logic into a single storage array system rather than distributing complexity across multiple hosts. This consolidation simplifies configuration while maintaining reliability.

Inventive Principle:

Principle #5Merging (Combining)

3Reliability

If port failover is managed at the host level, then failure recovery is achieved, but deployment becomes expensive and inconvenient

Engineering Contradiction:

Improvefailure recovery capabilityVSAvoiddeployment ease

Core Design Contradiction:

ReliabilityVSEase of manufacture

Solution Approach 1:

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

Inventive Principle:

Principle #24Intermediary (Mediator)

4Reliability

If requests are redirected from virtual ports to physical ports during failback, then original connectivity is restored, but request queuing and coordination overhead is introduced

Engineering Contradiction:

Improveconnectivity restorationVSAvoidrequest redirection delay

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The storage array pre-establishes virtual port mappings and failback coordination mechanisms before failures occur. This preliminary setup reduces the time required for request redirection during failback operations.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements coordination processing where storage processors exchange status information to determine when failback is appropriate. This feedback mechanism ensures requests are redirected at the optimal time to minimize delays.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS8909980B1Coordinating processing for request redirection

Publication Date: 2014.12.09 EMC IP HLDG CO LLC

AI summary

Described are techniques for coordinating processing to redirect requests. First and second storage processors of a data storage system are provided. Requests are directed to a first set of physical target ports of the first storage processor. The second storage processor is unavailable and has a second set of virtual target ports. Each virtual port of the second set is virtualized by a physical port of the first set. First processing is performed to redirect requests from the second set of virtual ports to a third set of physical ports of the second storage processor. First processing includes booting the second storage processor, directing requests to the third set of physical ports of the second storage processor rather than second set of virtual ports, and queuing requests received at the third set of physical ports until completion of pending requests previously received at the second set of virtual ports.