Slice File Recovery Using Dead Replicas After Storage Errors

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing technologies struggle to effectively recover a primary slice file from data corruption caused by storage device errors, especially when there are no live replica slice files available, leading to unacceptable data loss.

Innovation Solution

The technology employs an automated process using dead replica slice files to recover the primary slice file by programmatically identifying and utilizing up-to-date, non-stale data from dead replica slice files through checksum comparisons, allowing for the repair of corrupted blocks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If replica slice files are maintained as live copies for recovery, then data recovery capability is improved, but storage space consumption increases

Engineering Contradiction:

Improvedata recovery capabilityVSAvoidstorage space consumption

Core Design Contradiction:

ReliabilityVSVolume of stationary object

Solution Approach 1:

The patent implements a mechanism to identify and discard stale replica slice files that no longer contain useful data for recovery. By using checksum comparisons between primary and replica slice files, the system automatically determines which replicas are stale and excludes them from recovery operations, thereby reducing unnecessary storage space consumption while maintaining effective recovery capability from valid replicas

Inventive Principle:

Principle #34Discarding and recovering

Solution Approach 2:

The system creates multiple replica slice files as copies of the primary slice file for recovery purposes. These replicas are distributed across different storage locations and are maintained to enable recovery operations. The checksum-based identification system allows the system to manage these copies efficiently by knowing which ones are valid and which are stale

Inventive Principle:

Principle #26Copying

2Ease of operation

If manual intervention is used for slice file recovery, then operational control is improved, but recovery time and complexity increase

Engineering Contradiction:

Improveoperational controlVSAvoidrecovery time

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The patent implements an automated recovery mechanism that performs slice file recovery without requiring manual intervention. The system automatically identifies corrupted slice files, locates valid replica files using checksum comparisons, and executes recovery operations autonomously. This self-service approach reduces recovery time significantly while providing operators with visibility through status information

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system incorporates feedback mechanisms through status information that operators can view to monitor recovery progress. The automated process provides feedback about which slice files have been recovered, which are still processing, and the overall health of the storage system, allowing operators to understand the recovery state without direct intervention

Inventive Principle:

Principle #23Feedback

3Productivity

If all replica slice files are used for recovery without validation, then recovery speed is improved, but data integrity deteriorates

Engineering Contradiction:

Improverecovery speedVSAvoiddata integrity

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The patent performs preliminary validation of replica slice files using checksum comparisons before using them for recovery. The system proactively identifies which replicas are stale by comparing their checksums against the primary slice file's checksum, and excludes these invalid replicas from recovery operations. This preliminary action ensures data integrity is maintained while still allowing rapid recovery from valid replicas

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20250362814A1Slice file recovery using dead replica slice files

Publication Date: 2025.11.27 NETAPP INC

AI summary

Techniques are provided for repairing a primary slice file, affected by a storage device error, by using one or more dead replica slice files. The primary slice file is used by a node of a distributed storage architecture as an indirection layer between storage containers (e.g., a volume or LUN) and physical storage where data is physically stored. To improve resiliency of the distributed storage architecture, changes to the primary slice file are replicated to replica slice files hosted by other nodes. If a replica slice file falls out of sync with the primary slice file, then the replica slice file is considered dead (out of sync) and could potentially comprise stale data. If a storage device error affects blocks storing data of the primary slice file, then the techniques provided herein can repair the primary slice file using non-stale data from one or more dead replica slice files.