Data Recovery for Failed Memory Device Arrays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Parity-based recovery methods are impractical for reconstructing lost data in large memory device arrays due to performance degradation and lengthy reconstruction times, especially when a memory device fails, as they require reading multiple blocks and calculating parity, which is computationally expensive and time-consuming.
Innovation Solution
A data storage system utilizing two data storage devices, where one device stores multiple copies of data and the other uses parity information to recover lost data, allowing for efficient recovery by reading copied data from a second storage device and reconstructing data from parity blocks in the original device, thereby reducing reconstruction time and improving system performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If parity-based recovery is used to reconstruct lost data from a failed memory device, then data reliability is maintained, but system performance degrades and reconstruction time becomes excessively long
Solution Approach 1:
The patent segments the data recovery process into two distinct phases: (1) rapid initial recovery of commonly accessed data blocks from a second storage device using copied data, and (2) subsequent recovery of remaining data blocks from parity information. This segmentation allows the system to prioritize performance-critical operations while maintaining complete data reliability through the fallback parity-based recovery mechanism.
Solution Approach 2:
The patent implements preliminary action by pre-copying frequently accessed data blocks to a second storage device before failure occurs. When a memory device fails, these pre-positioned copies enable immediate recovery of hot data without requiring time-consuming parity calculations, thus maintaining system performance while preserving data reliability.
2Reliability
If parity-based recovery is used to reconstruct lost data from a failed memory device, then data reliability is maintained, but reconstruction time becomes excessively long
Solution Approach 1:
The recovery process is divided into time-sensitive and time-insensitive segments. Time-critical data blocks are recovered immediately from the second storage device using simple read operations, while non-critical blocks are recovered later using parity information. This segmentation dramatically reduces the perceived reconstruction time while ensuring complete data reliability.
Solution Approach 2:
The patent applies partial action by recovering only the most critical data blocks immediately from the second storage device, rather than waiting to recover all blocks through the slower parity-based method. This partial recovery approach reduces reconstruction time significantly while maintaining data reliability through subsequent complete recovery.
3Loss of information
If parity-based recovery is used to reconstruct lost data, then data can be recovered, but computational cost and time consumption increase due to reading multiple blocks and calculating parity
Solution Approach 1:
The patent creates and maintains a copy of frequently accessed data blocks on a second storage device. When recovery is needed, the system reads directly from this copy rather than performing computationally expensive parity calculations, thereby reducing energy consumption while maintaining full data recovery capability for the copied blocks.
Solution Approach 2:
The system performs preliminary copying of data blocks to a second storage device before failure occurs. This pre-positioning of data eliminates the need for energy-intensive parity calculations during recovery operations, reducing computational cost while preserving the ability to recover all lost data through the combination of copies and parity information.
Data Source
AI summary
Some aspects of the disclosure relate to a data storage system that includes multiple memory device storage devices. If a memory device of a memory device array fails within a first data storage device, some portions of the lost or corrupted data from the failed memory device are recovered by reading them from a second data storage device. Other portions of the lost or corrupted data from the failed memory device are recovered from parity information in the first data storage device.