SSD Mapping Audit for Verified Secure Erase
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a demand for securely and reliably erasing specific data stored in nonvolatile memory systems like SSDs using a physical secure erase method and verifying the correctness of the erase operation.
Innovation Solution
A memory system with a controller that manages mapping information between logical and physical addresses, performs secure erase operations, and transitions to an Audit mode to confirm the correctness of the erase by reading data from a stored mapping information copy, ensuring all secure erase target data is physically erased.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If physical secure erase method is used to erase data from SSD, then data security is improved, but confirmation of erase correctness becomes difficult
Solution Approach 1:
The patent implements a feedback mechanism by storing mapping information before secure erase and using it to verify erase correctness. The controller reads the stored mapping information after erase operation and confirms whether the data at the mapped physical address has been successfully erased, providing feedback on erase correctness to the host.
Solution Approach 2:
The patent creates a copy of the mapping information that maps logical addresses to physical addresses before executing the secure erase operation. This copy is stored in non-volatile memory and later used to verify that the data at the corresponding physical addresses has been properly erased, enabling confirmation of erase correctness without compromising security.
2Reliability
If secure erase operation is executed on mixed data from multiple namespaces, then data security is improved, but operation complexity increases
Solution Approach 1:
The patent segments the secure erase operation into distinct phases: first identifying blocks containing target data by reading mapping information, then executing erase operations on those specific blocks. This segmentation allows the system to handle mixed data from multiple namespaces by processing only the relevant blocks marked in the mapping information, reducing overall complexity.
Solution Approach 2:
The patent uses mapping information as an intermediary between the secure erase command and the actual data erase operation. The mapping information serves as a mediator that identifies which physical blocks contain the logical data to be erased, allowing the controller to accurately target and erase only the necessary data while ignoring other namespaces' data in the same blocks.
3Reliability
If mapping information is stored before secure erase, then verification capability is improved, but storage space requirement increases
Solution Approach 1:
The patent applies local quality by storing mapping information selectively only for the logical blocks that need to be verified after secure erase, rather than storing complete mapping information for the entire drive. The controller identifies which mapping entries correspond to the target data and stores only those, reducing the storage space requirement while maintaining verification capability for the erased data.
Data Source
AI summary
According to one embodiment, a controller, in response to receiving, from a host, a first command requesting secure erase of secure erase target data associated with a first logical area identifier, stores a copy of first mapping information that corresponds to the first logical area identifier, among mapping information that is included in a first table. The controller executes at least a data erase operation for one or more first blocks storing the secure erase target data. In a first mode, the controller, in response to receiving, from the host, a read command that specifies the first logical area identifier, reads data from a storage location corresponding to a first physical address that is mapped to the first logical area identifier in the copy of the first mapping information.