Fail-Safe Storage Device Inode Deletion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing file systems such as FAT and exFAT lack mechanisms to ensure fail-safe operations during unexpected power loss or storage device disconnections, leading to data corruption issues like cross-linked clusters, lost clusters, and corrupt directories, without the feasibility of adding journaling due to compatibility concerns and performance penalties.
Innovation Solution
The techniques involve marking directory entries as not in use and subsequently freeing associated clusters, allocating and deallocating clusters in a manner that ensures data integrity and consistency without metadata duplication, thereby preventing data loss and corruption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If journaling mechanisms are added to FAT and exFAT file systems to ensure fail-safe operations, then data integrity and reliability are improved, but device complexity increases and performance penalties occur
Solution Approach 1:
The patent extracts the essential fail-safe functionality from complex journaling mechanisms and implements it through a simplified two-stage deletion process. Instead of adding a full journaling subsystem, the invention isolates and implements only the critical metadata update sequence (marking directory entries as deleted, then updating allocation bitmap, then freeing clusters) to achieve reliability without the overhead of complete journaling infrastructure.
Solution Approach 2:
The patent applies local quality by implementing fail-safe operations only where critically needed - specifically in the metadata update sequence during deletion operations. Rather than making the entire file system complex and journaled, the invention selectively enhances only the vulnerable deletion path with ordered updates, leaving the rest of the FAT/exFAT structure simple and compatible.
2Reliability
If journaling is implemented to prevent data corruption during power loss or disconnection, then reliability is improved, but power consumption and write amplification increase
Solution Approach 1:
The patent applies preliminary action by pre-establishing a defined update sequence for metadata operations before failures can occur. The deletion process is structured to first mark directory entries as deleted, then update the allocation bitmap, and finally free clusters - ensuring that at any failure point, the file system remains in a consistent state without requiring additional journaling writes that would consume extra power.
Solution Approach 2:
The patent implements partial action by applying fail-safe mechanisms only to the specific metadata update sequence needed for deletion operations, rather than implementing comprehensive journaling for all file system operations. This selective approach provides sufficient reliability for the critical deletion path while minimizing unnecessary write operations and power consumption.
3Productivity
If traditional deletion methods are used where clusters are freed immediately, then productivity is improved, but data corruption risks increase during unexpected failures
Solution Approach 1:
The patent applies preliminary action by performing directory entry marking as the first step in deletion, establishing a safe state before cluster freeing occurs. This preliminary update ensures that even if failure occurs during cluster release, the directory already reflects the deleted state, preventing corruption while maintaining efficient deletion throughput through the streamlined two-stage process.
Data Source
AI summary
Techniques for enabling fail-safe operations of storage devices are described. A device may be configured to delete an inode in a failsafe manner. A device may be configured to mark directory entries as not in use and ensure this goes to disk before anything else. A device may be configured to create an inode in a failsafe manner. A device may be configured to allocate and initialize a cluster and ensure this is written to disk before anything else. In this manner, a volume may be in a predictable state upon a failure occurring.