Dual Active Storage Controller Path Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In high availability storage configurations with redundant dual active controllers, host systems often send commands to storage controllers that do not have access to the data volume, leading to errors and delays due to the visibility of all data paths from both controllers, causing I/O performance penalties and high availability feature loss.

Innovation Solution

The solution involves identifying the owned logical unit for each controller and only reporting the logical unit number of the owned logical unit, making the redundant dual active controllers appear as two independent storage targets to the host system, thus optimizing data paths and avoiding I/O performance penalties by ensuring only optimized data paths are used.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If all data paths from both redundant controllers are visible to the host system, then high availability and redundancy are provided, but I/O errors and performance penalties occur due to commands being sent to controllers without access to the data volume

Engineering Contradiction:
Improvehigh availabilityVSAvoidI/O performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent segments the visibility of data paths by controller, where each controller presents only its owned logical units to the host system. This segmentation prevents the host from sending commands to the wrong controller while maintaining redundancy, as each controller independently manages its subset of data paths and logical units.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary mechanism (the controller's selective reporting of logical units) between the host system and the data paths. This intermediary filters which data paths are visible to the host based on controller ownership, preventing direct conflicts while preserving the underlying redundant architecture.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If all data paths are visible from both controllers, then redundancy is maintained, but boot time increases due to volume thrashing and I/O errors

Engineering Contradiction:
ImproveredundancyVSAvoidboot time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

During system initialization, each controller segments its visibility to logical units, reporting only those it owns. This prevents volume thrashing during boot where the system would otherwise attempt to access volumes through both controllers, significantly reducing boot time while preserving redundancy capabilities.

Inventive Principle:
Principle #1Segmentation

3Reliability

If the storage array uses vendor-specific configurations for multiple-path solutions, then high availability can be maintained, but device complexity and configuration difficulty increase

Engineering Contradiction:
Improvehigh availabilityVSAvoidconfiguration complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements a universal approach where the storage array itself provides the intelligence to manage multiple paths correctly. Each controller autonomously determines which logical units to report, making the system compatible with generic multiple-path solutions from any vendor without requiring proprietary configurations or software.

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

Data Source

PatentUS7847801B2Method for reporting redundant controllers as independent storage entities
Publication Date: 2010.12.07 NETAPP INC
  • US7847801B2 patent drawing
  • US7847801B2 patent drawing
  • US7847801B2 patent drawing

AI summary

A computer implemented method, apparatus, and computer usable program code are provided for managing dual active controllers in a high availability storage configuration. Redundant dual active controllers in high availability storage configurations are made to appear as individual storage target devices to a host system. Each controller owns certain volumes of data storage. When a host system sends a request to identify available data volumes, the controller that owns certain volumes provides preferred paths to those owned volumes. The host system may also send an inquiry to a controller that asks the controller about data volumes not owned by the controller. For such inquiries, no paths to the non-owned data volumes are returned to the host system.