Memory Controller Shadow Tracker for Secure NVM Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current secure non-volatile memory systems face challenges in achieving ultra-low overhead and rapid recovery times due to the need to persist security metadata, particularly with Merkle Trees and encryption counters, which are not effectively addressed by existing solutions that incur high overhead or prolonged recovery times.
Innovation Solution
A system and method utilizing a memory controller that persistently tracks addresses of blocks in a shadow tracker memory region of the non-volatile memory, allowing for quick recovery of security metadata by rebuilding only affected parts of the Merkle Tree and counter caches, thereby reducing recovery time and overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If security metadata is persistently stored in NVM to ensure crash consistency, then reliability is improved, but device complexity and overhead increase
Solution Approach 1:
The patent extracts only the essential information needed for recovery (addresses of modified metadata blocks) and stores it in the shadow tracker, rather than persisting the entire Merkle tree and all metadata. This selective extraction reduces complexity while maintaining crash consistency for the critical recovery path.
Solution Approach 2:
The shadow tracker is updated in advance during normal operation to record addresses of modified metadata blocks before a crash occurs. This preliminary tracking ensures that when a crash happens, the recovery process only needs to rebuild the specific affected parts rather than scanning the entire tree, thus reducing complexity without sacrificing reliability.
2Reliability
If all levels of Merkle Tree are updated atomically on each counter update to ensure consistency, then reliability is improved, but use of energy and write operations increase
Solution Approach 1:
The patent segments the Merkle tree updates by tracking only the affected paths from leaf to root in the shadow tracker, rather than updating all levels atomically. This segmentation allows the system to maintain consistency information for recovery while avoiding the energy-intensive full-tree updates on every counter modification.
Solution Approach 2:
Instead of performing complete atomic updates of all Merkle tree levels, the patent applies partial action by only tracking and persisting the addresses of modified blocks in the shadow tracker. This partial persistence approach maintains sufficient reliability for recovery while dramatically reducing the energy consumption and write operations required.
3Reliability
If complete recovery of Merkle Tree is performed after crash to ensure integrity, then reliability is improved, but loss of time increases
Solution Approach 1:
The patent extracts only the critical recovery information (addresses of modified metadata blocks) from the entire Merkle tree and stores it in the shadow tracker. During recovery, only these extracted addresses are used to rebuild the affected parts of the tree, eliminating the need to scan and reconstruct the entire tree and thus dramatically reducing recovery time while maintaining integrity.
Solution Approach 2:
The shadow tracker performs preliminary recording of modified block addresses during normal operation. This advance preparation ensures that during crash recovery, the system only needs to rebuild the specific affected portions of the Merkle tree rather than the entire structure, reducing recovery time from hours to seconds while preserving integrity.
4Speed
If metadata are cached in volatile cache structures to improve access speed, then speed is improved, but loss of information occurs on power loss or crash
Solution Approach 1:
The shadow tracker acts as an intermediary between the volatile cache and the non-volatile NVM storage. It captures the addresses of modified metadata blocks from the volatile cache and persists this information to NVM. This intermediary mechanism allows the system to maintain fast cache access during normal operation while ensuring that critical recovery information is preserved in non-volatile storage after crashes or power losses.
Data Source
AI summary
A device comprising a memory controller coupled to a non-volatile memory (NVM) device with a shadow tracker memory region. The controller comprises a low-overhead and low recovery time for integrity-protected systems by recovering a secure metadata cache. The controller is configured to persistently track addresses of blocks in the secure metadata cache in the NVM device when a miss occurs, and track the persistent addresses, after the miss. The controller is configured to rebuild affected parts of the secure metadata cache associated with the persistent addresses in the NVM device. A system is provided which includes the memory controller interfaced with an NVM device with the shadow tracker memory region.