Disk Array Control Device Detecting Unfinished-Writing Failures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current disk array devices face challenges in identifying and recovering from 'unfinished-writing failures' where data is not properly written to magnetic disks, leading to data loss and invalidation of other disks' data due to fraudulent or incomplete writes, which existing RAID technologies struggle to detect and correct.
Innovation Solution
A disk array control method that calculates recovery data using a majority logic to identify storage devices experiencing unfinished-writing failures by detecting incoherency in redundant configurations without abnormality reports, allowing for correct data recovery and validation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If RAID technology using data redundancy is employed to recover data from damaged magnetic disks, then data recovery capability is improved, but the ability to detect which specific magnetic disk has failed is lost
Solution Approach 1:
The patent applies preliminary action by writing verification data (pattern data) to magnetic disks before actual data storage operations. This verification data serves as a reference marker that allows the system to detect whether a disk has properly received and stored data. When a failure occurs, the verification data comparison reveals which specific disk failed, solving the detection precision problem while maintaining RAID recovery capabilities.
Solution Approach 2:
The patent implements feedback by continuously comparing verification data read from magnetic disks with the expected pattern data. This feedback mechanism provides real-time information about disk status and data integrity. When mismatch is detected, the system receives feedback about which specific disk has failed, enabling precise failure detection while maintaining the overall RAID redundancy structure for data recovery.
2Productivity
If magnetic disks are monitored for writing completion reports, then writing status tracking is improved, but unfinished-writing failures remain undetected when disks report completion before actual writing occurs
Solution Approach 1:
The patent applies preliminary action by pre-writing verification pattern data to specific locations on magnetic disks before the main data writing process. This verification data acts as a sentinel that allows the system to verify whether the disk actually completed the writing operation. Even if a disk falsely reports writing completion, the verification data comparison will reveal the discrepancy, thus solving the reliability problem without reducing tracking efficiency.
Solution Approach 2:
The patent introduces verification data as an intermediary element between the writing process and the completion confirmation. This intermediary verification data serves as proof of actual writing completion. The system uses this intermediary marker to validate whether a disk truly completed writing before considering the operation successful, thereby detecting fraudulent completion reports while maintaining efficient writing status tracking.
Data Source
AI summary
In a disk array configured to distribute and store a plurality of data on a plurality of storage devices, a disk array control device is configured to calculate recovery data for a plurality of storage devices upon detecting an incoherency in a redundant configuration without receiving an abnormality report from a storage device having written predetermined data thereon and to identify the storage device, relating to the correct recovery data determined according to a majority logic, as a storage device inviting an unfinished-writing failure.