Storage Cache Re-silvering via Mirrored Data Recovery

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In storage management, when a storage device becomes unavailable, bringing it back online is complicated by stale mirrored data that does not reflect recent updates, and existing methods for re-silvering (making stale data current) are inefficient and risk data loss due to cache failures.

Innovation Solution

Implementing a write-back mode in storage devices that use non-volatile cache and cache metadata to efficiently re-silver data by identifying dirty cache copies and using a mirrored, up-to-date version from another storage device to update primary storage, ensuring data availability and integrity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If data is stored in cache to provide quick access, then access speed is improved, but data loss risk increases when cache fails

Engineering Contradiction:
Improvedata access speedVSAvoiddata loss risk
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The system establishes a mirror copy mechanism beforehand, creating a redundant data backup on a separate storage device. When cache failure occurs, the mirror copy serves as a protective cushion to prevent data loss, allowing the system to recover without losing the cached data.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

2Reliability

If data mirroring is used to ensure availability, then reliability is improved, but re-silvering efficiency deteriorates due to stale data

Engineering Contradiction:
Improvedata availabilityVSAvoidre-silvering efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary actions by maintaining a mirror copy of the cache data on a separate storage device in real-time. When the storage device becomes unavailable, the mirror copy is already prepared and can be quickly copied back to restore data, eliminating the need for time-consuming re-silvering operations.

Inventive Principle:
Principle #10Preliminary action

3Loss of information

If complete re-silvering is performed to update stale data, then data currentness is improved, but time consumption increases

Engineering Contradiction:
Improvedata currentnessVSAvoidre-silvering time
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

Instead of performing complete re-silvering of all data, the system uses the pre-existing mirror copy to selectively restore only the necessary cached data that was lost. This partial action approach restores data currentness without the time penalty of copying entire datasets.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS9063908B2Rapid recovery from loss of storage device cache
Publication Date: 2015.06.23 ORACLE INT CORP
  • US9063908B2 patent drawing
  • US9063908B2 patent drawing
  • US9063908B2 patent drawing

AI summary

Dirty data in a storage device is made current through rapid re-silvering, which uses a mirrored and up-to-date version of the dirty data from another storage device to recover the data. Because under rapid re-silvering cache metadata in volatile memory survives the failure of the cache, the cache metadata is used to determine which subset of data from the other storage device needs to be copied to the storage device being re-silvered. During re-silvering, cache metadata is used to determine which I/O requests from clients are requests for data that is not stale.