RAID-DP Lost Write Detection via Parity Scrubbing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage systems, particularly those using RAID with multiple parity, are inadequate in detecting and correcting 'lost writes' due to faulty hardware, as existing error detection techniques fail to identify corrupted blocks in such scenarios, leading to potential data loss and performance degradation.
Innovation Solution
A method is introduced that uses a redundancy scheme capable of reconstructing data in X or fewer mass storage devices, allowing for the identification of specific disks containing erroneous data, even in the presence of lost writes, by employing RAID-DP with row-diagonal parity, enabling detection and correction during parity scrubbing without impacting normal read and write operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If block-appended checksum is used for error detection, then corruption due to bit flips, partial writes, sector shifts and block shifts can be detected, but corruption due to lost block writes cannot be detected
Solution Approach 1:
The patent combines block-appended checksum with RAID parity mechanisms to create a hybrid error detection system. The checksum detects corruption within blocks while the parity mechanism detects lost writes at the stripe level, achieving comprehensive error detection by merging these two complementary approaches.
Solution Approach 2:
The patent extends the functionality of RAID parity beyond its traditional error correction role to also serve as a lost write detection mechanism. By analyzing parity mismatches during read operations, the system universally detects both corruption errors and lost write errors using the same parity infrastructure.
2Ease of repair
If single parity schemes such as RAID-3, RAID-4 or RAID-5 are used, then a single bad block can be reconstructed from parity and remaining data blocks, but there is not enough information to determine which disk contains the corrupted block
Solution Approach 1:
The patent implements a feedback mechanism where read operations trigger parity verification. When parity mismatches are detected, the system provides feedback to identify which specific disk contains the corrupted block by comparing expected parity values with actual values, enabling precise error location without requiring additional hardware.
Solution Approach 2:
The patent segments the error detection process into distinct phases: parity calculation, parity comparison, and error location identification. By dividing the detection process into these manageable segments, the system efficiently locates corrupted blocks without overwhelming computational overhead.
3Reliability
If mirroring data to two disks is used, then mismatch between two copies can indicate lost writes, but it is not possible to determine which data are correct
Solution Approach 1:
The patent introduces parity information as an intermediary reference to resolve conflicts between mirrored data copies. When mismatches occur between mirrored blocks, the parity value serves as a mediator to determine which copy is correct, eliminating the ambiguity inherent in simple mirroring approaches.
4Reliability
If parity checking is performed continuously to detect bad blocks, then lost writes can be detected, but computing resources are heavily consumed and performance degrades
Solution Approach 1:
The patent implements periodic parity checking during scheduled maintenance windows or idle periods rather than continuous checking during normal operations. This periodic approach maintains data integrity through regular verification while minimizing performance impact by avoiding constant parity calculations during active read/write operations.
Solution Approach 2:
The patent performs parity verification in advance during write operations or idle periods before data corruption can impact system operations. By proactively checking parity and identifying lost writes before they cause failures, the system maintains high reliability without requiring continuous monitoring that would degrade performance.
Data Source
AI summary
A storage server stores data in a redundancy group of mass storage devices according to a data redundancy scheme. The redundancy scheme enables reconstruction of all data stored in the redundancy group if N or fewer mass storage devices in the redundancy group contain erroneous data, where N is at least 2. In response to determining that the redundancy group contains erroneous data, N−1 mass storage devices in the redundancy group can be specifically identified as containing the erroneous data.