Disk Failure Handling via Sector Set Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional disk failure determining and handling techniques are insufficient in granularity, failing to support fine-grained processing and meeting user needs, as they focus on entire disk failures rather than sector-level issues, leading to inefficient data protection and increased CPU and I/O costs.
Innovation Solution
A method that determines sector set failure types by analyzing disk health at a fine granularity, using a failure determining model to predict sector set failures and perform targeted handling actions such as data migration or recovery based on threshold numbers, allowing for efficient and cost-effective disk failure management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional disk failure determining and handling techniques focus on entire disk failures, then unified processing can be performed on the entire disk, but the granularity is insufficient and cannot support fine-grained processing of disks
Solution Approach 1:
The patent divides the disk into multiple sector sets and performs failure detection and handling at the sector set level rather than treating the entire disk as a single unit. This segmentation enables fine-grained processing where each sector set can be independently monitored and managed, resolving the contradiction between detection precision and processing complexity by operating on manageable subdivisions of the disk
Solution Approach 2:
The patent applies different handling strategies to different sector sets based on their specific failure conditions. Sector sets with fewer failures undergo recovery operations while those exceeding thresholds trigger data migration or disk replacement. This localized quality approach allows precise failure handling tailored to each sector set's condition, achieving high detection granularity without overwhelming system complexity
2Productivity
If conventional techniques perform unified processing on the entire disk when failure is detected, then simple handling can be performed, but data protection efficiency is reduced and CPU and I/O costs increase
Solution Approach 1:
By segmenting the disk into sector sets and independently processing each sector set based on its failure condition, the patent avoids unnecessary processing of healthy sectors. This selective processing improves data protection efficiency by focusing resources only on affected areas while reducing overall CPU and I/O costs compared to unified disk-level processing
Solution Approach 2:
The patent applies partial action by performing recovery operations only on sector sets with fewer failures and data migration only on those exceeding the threshold, rather than processing the entire disk uniformly. This partial processing approach optimizes resource utilization by applying the minimum necessary action to each sector set, improving productivity while minimizing energy loss
Data Source
AI summary
Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for failure handling. This failure handling method includes determining a sector set failure type associated with at least one failed sector set of a disk; if the sector set failure type indicates that the number of failed sector sets in the at least one failed sector set is greater than a first threshold number, generating an instruction for replacing the disk; and otherwise performing at least one of the following: migrating data from a failed sector set in which the number of failed sectors is greater than a second threshold number to a spare sector set, and performing a failure recovery for a failed sector set in which the number of failed sectors is less than or equal to the second threshold number.