Logical Block Key Tables for Fast Persistent Memory Encryption
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data encryption methodologies on memory devices face challenges in managing encryption keys per logical block, leading to inefficiencies in key identification and access, which impact device performance.
Innovation Solution
A memory sub-system manages encryption keys per logical block using key tables with key tags and identifiers, and employs a key cache for rapid access, ensuring correct key usage and minimizing performance impact.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If encryption keys are managed per logical block on a persistent memory device, then data security is improved, but key identification and access efficiency deteriorates
Solution Approach 1:
The patent segments the key management system into multiple key tables (first key table in local memory, second key table in RAM) organized by key tags corresponding to different logical blocks. This segmentation allows efficient routing and access to specific keys without scanning all keys, resolving the contradiction between per-block security and access efficiency.
Solution Approach 2:
The patent performs preliminary organization of encryption keys into structured key tables with key tags and identifiers before actual data operations. The key cache pre-loads and stores frequently accessed keys, so when a data operation occurs, the correct key is already available, eliminating delays in key identification and access.
2Speed
If multiple key tables are used for encryption key management, then key access speed is improved, but device complexity increases
Solution Approach 1:
The patent assigns different key tables to different memory locations with specialized functions: local memory contains a first key table for frequently accessed keys, while RAM contains a second key table for less frequently accessed keys. This local quality differentiation optimizes access speed for different key types while managing complexity through functional specialization.
Solution Approach 2:
The patent introduces key tags as intermediary identifiers that link logical blocks to their corresponding encryption keys in the key tables. This intermediary layer simplifies the access process by providing a direct lookup mechanism, reducing the complexity of managing multiple keys while maintaining high access speed.
Data Source
AI summary
A command to perform a data operation at a memory device is received. The command includes an encryption key tag. A first key table is accessed from local memory. The first key table includes a first set of key entries corresponding to a first set of encryption keys. The first key table is searched to determine whether it includes an entry corresponding to the encryption key tag. Based on determining the first key table does not include an entry corresponding to the tag, a second key table is accessed from RAM. The second key table includes a second set of key entries corresponding to a second set of encryption keys. A key entry corresponding to the encryption key tag is identified from the second key table. The key entry includes an encryption key corresponding to the encryption key tag. The command is processed using the encryption key.