Drive Array Self-Healing via Deletable Stripe Zones

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Data storage systems face challenges in efficiently recovering from partial or full drive failures, particularly when insufficient spare capacity is provisioned, leading to potential data loss and performance degradation, as existing RAID configurations may not effectively utilize user storage space for self-healing without compromising user-accessible storage capacity.

Innovation Solution

A drive array is configured to map stripe zones across drives, allowing for the identification and offline use of deletable stripe zones for rebuilding affected areas, even if they contain user data, to reclaim storage space and maintain system integrity during drive failures, thereby enabling self-healing without requiring user input.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If spare capacity is provisioned for drive failure recovery, then system reliability is improved, but user-accessible storage capacity is reduced

Engineering Contradiction:

Improvedrive failure recovery capabilityVSAvoiduser-accessible storage capacity

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The system dynamically converts user data stripe zones to spare capacity only when drive failure occurs, rather than statically allocating spare capacity. This allows the storage system to adapt between maximizing user capacity during normal operation and providing recovery capability during failure events.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the functional parameter of existing stripe zones from 'user data storage' to 'recovery spare capacity' when needed. This parameter transformation allows the same physical storage space to serve different purposes at different times, resolving the contradiction between capacity and reliability.

Inventive Principle:

Principle #35Parameter changes

2Extent of automation

If deletable stripe zones are taken offline for rebuilding, then system self-healing is enabled, but user storage availability is temporarily reduced

Engineering Contradiction:

Improveself-healing capabilityVSAvoiduser storage availability

Core Design Contradiction:

Extent of automationVSProductivity

Solution Approach 1:

The storage system performs self-healing by automatically identifying failed drives, selecting appropriate stripe zones for recovery, and executing the rebuild process without external intervention. This automation enables the system to restore itself while minimizing impact on users.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system takes only the minimum necessary stripe zones offline for rebuilding rather than the entire array. By selectively offline-ing only the specific zones needed for recovery, the system enables self-healing while preserving most user storage availability.

Inventive Principle:

Principle #16Partial or excessive action

3Adaptability or versatility

If stripe zone capacity is kept small to fit on average drive sizes, then drive replacement flexibility is improved, but recovery speed is reduced

Engineering Contradiction:

Improvedrive replacement flexibilityVSAvoidrecovery speed

Core Design Contradiction:

Adaptability or versatilityVSSpeed

Solution Approach 1:

The system divides the array into multiple stripe zones that can be independently managed and rebuilt. This segmentation allows parallel rebuilding of multiple zones simultaneously, increasing overall recovery speed while maintaining the flexibility to work with standard drive sizes.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS10877844B2Using deletable user data storage space to recover from drive array failure

Publication Date: 2020.12.29 SEAGATE TECH LLC

AI summary

A plurality of stripe zones are mapped across drives in an array. A capacity of each of the stripe zones is smaller than or equal to an average size of the drives. A failed drive is determined in the array. A deletable stripe zone is selected that is being used for user data storage. The deletable stripe zone is taken offline and used to rebuild a subset of the plurality of the stripe zones affected by the failed drive.