At Risk Data Caching for RAID Degraded Conditions

Solution Overview

Problem

RAID arrays face challenges in quickly restoring data without data loss during degraded conditions, such as drive failures or media errors, especially when multiple data portions become invalid, leading to uncorrectable conditions due to coherency errors.

Innovation Solution

A computer-implemented method that senses degraded conditions in a RAID array by reconstructing an invalid data portion using valid data portions and parity, storing it in a mirrored cache memory system, updating the parity, and writing the updated content back to the array, ensuring data integrity and availability across multiple storage processors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data is stored in a RAID array with parity portions, then data redundancy and fault tolerance are improved, but data coherence and integrity may deteriorate during degraded conditions due to coherency errors

Engineering Contradiction:

Improvedata fault toleranceVSAvoiddata coherence

Core Design Contradiction:

ReliabilityVSStability of the object's composition

Solution Approach 1:

The system performs preliminary reconstruction of invalid data portions using valid data portions and parity information before writing updated content. By pre-reconstructing the invalid data portion and storing it in a mirrored cache memory system, the system ensures data coherence is maintained even during degraded conditions, preventing coherency errors from occurring in the first place

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

A mirrored cache memory system is introduced as an intermediary between the RAID array and the invalid data portions. This cache system stores reconstructed data portions and provides a buffer that maintains data coherence, allowing the system to handle degraded conditions without losing data integrity or creating coherency errors

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If multiple data portions become invalid in a RAID array, then data loss risk increases, but restoration time may be extended due to complex reconstruction processes

Engineering Contradiction:

Improvedata availabilityVSAvoidrestoration time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary reconstruction of invalid data portions using valid data portions and parity information before writing updated content. By pre-reconstructing the invalid data portion and storing it in a mirrored cache memory system, the system ensures data coherence is maintained even during degraded conditions, preventing coherency errors from occurring in the first place

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system automatically detects degraded conditions and triggers the reconstruction process without external intervention. The storage processor autonomously identifies invalid data portions, reconstructs them using available valid portions and parity, and updates the array, enabling self-healing that reduces restoration time and maintains data availability

Inventive Principle:

Principle #25Self-service

3Manufacturing precision

If real-time data reconstruction is performed in a degraded RAID array, then data integrity is improved, but system complexity increases due to additional processing steps

Engineering Contradiction:

Improvedata integrityVSAvoidsystem complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

A mirrored cache memory system is introduced as an intermediary between the RAID array and the invalid data portions. This cache system stores reconstructed data portions and provides a buffer that maintains data coherence, allowing the system to handle degraded conditions without losing data integrity or creating coherency errors

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system creates a copy of the reconstructed data portion and stores it in the mirrored cache memory system. This copying mechanism preserves the original data structure and simplifies the reconstruction process by working with copies rather than directly modifying the original array, thereby maintaining data integrity while managing system complexity

Inventive Principle:

Principle #26Copying

Data Source

PatentUS9405626B1At risk data caching (ARDC)

Publication Date: 2016.08.02 EMC IP HLDG CO LLC

AI summary

A method, computer program product, and computing system for sensing the occurrence of an degraded condition within a data stripe of a RAID array. The data strip includes one or more valid data portions, one invalid data portion, and a parity portion. A request to write updated content to a target data portion within the data stripe is received, wherein the target data portion is one of the valid data portions. The valid data portions and the parity portion are read from the data stripe. The invalid data portion is reconstructed using the valid data portions and the parity portion, thus generating a reconstructed data portion. The reconstructed data portion is stored within a mirrored cache memory system. The parity portion is updated to define the updated content, thus defining an updated parity portion. The updated parity portion is written to the data stripe and the updated content is written to the target data portion within the data stripe.