De-duplication Aware Secure Delete Mechanism
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In data de-duplication environments, secure delete operations face challenges as they are not properly handled by the de-duplication engine, leading to incomplete data erasure and inefficient disk writes, especially when disk blocks are referenced by multiple files, making it difficult to bypass the engine without risking data corruption or requiring separate storage.
Innovation Solution
A mechanism that makes the de-duplication engine aware of secure delete writes, allowing them to be performed directly on physical disk blocks without de-duplication processing, and creates copies of referenced blocks to maintain an ignore list and ensure secure deletion, thereby avoiding de-duplication processing on identified blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If secure delete operations are performed through the de-duplication engine, then storage efficiency is maintained, but data erasure completeness deteriorates because the engine treats secure delete writes as regular data writes and creates de-duplication references
Solution Approach 1:
The patent extracts secure delete operations from the normal de-duplication processing path by identifying them as a special operation type. When the de-duplication engine detects a secure delete write operation, it bypasses the standard de-duplication reference creation process and directly writes to physical blocks, ensuring complete data erasure without leaving residues that would persist through normal de-duplication referencing.
Solution Approach 2:
The patent applies different processing quality to different operation types within the de-duplication engine. Regular data writes undergo full de-duplication processing to maintain storage efficiency, while secure delete writes receive specialized handling that prioritizes complete erasure. This local differentiation allows the system to maintain both storage efficiency and secure erasure capabilities simultaneously.
2Productivity
If de-duplication processing is applied to secure delete writes, then storage utilization is optimized, but deletion efficiency deteriorates due to redundant processing and failed erasure of referenced blocks
Solution Approach 1:
The patent implements preliminary identification of secure delete operations before they enter the de-duplication processing pipeline. By detecting the special operation type in advance, the system can route these operations through a dedicated path that bypasses redundant de-duplication processing, thereby improving deletion efficiency and eliminating wasted processing cycles on operations that require complete physical erasure.
3Reliability
If the de-duplication engine processes secure delete writes as regular writes, then operational simplicity is maintained, but data security deteriorates because overwritten data can be recovered through de-duplication references
Solution Approach 1:
The patent uses operation type identification to distinguish secure delete writes from regular data writes, analogous to using different colors to differentiate purposes. By tagging or identifying secure delete operations with a special marker, the de-duplication engine can apply fundamentally different processing logic: standard de-duplication for regular writes and direct physical erasure for secure deletes, thereby ensuring data security while maintaining clear operational distinction.
Data Source
AI summary
A mechanism is provided in a data processing system for de-duplication aware secure delete. Responsive to receiving a secure delete request for a file, the mechanism identifies a list of file blocks to be securely deleted from a physical disk device. Responsive to determining at least one virtual block of another file refers to a given disk block corresponding to a file block in the list, the mechanism copies the given disk block to generate a copied disk block in the physical disk device and updates a pointer of the at least one virtual block to refer to the copied disk block. The mechanism writes a secure delete pattern for each file block in the list of file blocks to a disk block in the physical disk device without performing de-duplication processing.