Memory Management Device Secure Page Table Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Nonvolatile memory devices are vulnerable to data theft and falsification during power down, as they can be physically accessed and manipulated externally, posing a significant threat to information processing devices, especially those used outside or in insecure locations.
Innovation Solution
A memory management device with a secure page table tree and verifier tree structure is implemented, where a secure operating system is protected by verification, and data is stored in a secure internal memory to prevent unauthorized access and falsification, while non-secure data is stored in external memory without verification to improve throughput.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If verification is performed for all data stored in nonvolatile memory, then data security and integrity are improved, but processing time and system complexity increase
Solution Approach 1:
The patent applies different verification strategies to different data types: secure data (OS, verifiers, page tables) undergoes strict verification, while non-secure data (application data, user files) is stored without verification. This local differentiation of verification quality resolves the contradiction by applying verification only where necessary for security, reducing overall processing time while maintaining data integrity for critical systems.
Solution Approach 2:
The memory system is segmented into secure memory (internal memory with verification) and non-secure memory (external memory without verification). This segmentation allows the system to maintain high reliability for critical operating system data while avoiding unnecessary verification overhead for non-critical data, thus resolving the time-loss contradiction.
2Reliability
If verification is performed for all data stored in nonvolatile memory, then data security is improved, but device complexity increases
Solution Approach 1:
The patent implements local quality by providing full verification infrastructure only for secure data regions (OS, verifiers, page tables in internal memory) while omitting it for non-secure data regions (application data in external memory). This resolves the complexity contradiction by concentrating verification complexity only where security is critical.
Solution Approach 2:
The verification mechanism is extracted and applied selectively only to critical system data rather than being universally applied to all data. The verifier calculation unit and verification logic are concentrated in securing the operating system and page tables, while external storage operates without this complexity, thus reducing overall device complexity while maintaining essential security.
3Reliability
If all data is stored in secure internal memory with verification, then data security is improved, but memory access speed and throughput decrease
Solution Approach 1:
The patent segments the memory system into secure internal memory for verification-critical data and non-secure external memory for high-speed bulk storage. This segmentation allows the system to achieve both high security for OS data and high throughput for application data, resolving the contradiction between security and productivity.
Solution Approach 2:
Different quality levels of security are applied locally: internal memory provides full verification-based security for critical data, while external memory provides high-speed access for non-critical data. This local quality differentiation resolves the contradiction by optimizing each memory region for its specific purpose.
4Productivity
If non-secure data is stored in external memory without verification, then processing speed is improved, but vulnerability to data falsification increases
Solution Approach 1:
The patent implements a dynamic security model where the level of verification applied to data depends on its security classification. Secure data (OS, verifiers) receives full verification protection, while non-secure data (application data) uses faster unverified storage. This dynamic approach resolves the contradiction by adapting the security level to the data's criticality rather than applying a uniform security level to all data.
Data Source
AI summary
In one embodiment, a storage unit stores a table tree and verifier tree. The table tree includes parent and child tables. The verifier tree includes parent and child verifiers associated with the parent and child tables, respectively. The parent verifier is used for verifying the child table and child verifier. A device stores a secure table tree corresponded to the table tree and used for address translation and a secure verifier tree corresponded to the verifier tree, to a secure storage unit. The device executes verification, based on verification information calculated based on a first child table and first child verifier in the storage unit and a first parent verifier in the secure verifier tree. The device sets the second address of the secure table tree such that the second address designates data in the first storage unit.