Dynamic Metablock Relinking for Non-Volatile Memory Health
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-volatile data storage devices face performance issues and reduced lifespan due to static metablock configurations, where failure of one block renders the entire metablock unusable, requiring resource-intensive data recovery and transfer.
Innovation Solution
Implementing a dynamic relinking mechanism that identifies metablocks with significant health differences, assigns them low write priority, and groups blocks with similar health values to form updated metablocks, extending their average useful life.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If blocks are grouped into static metablocks for fast write performance, then write speed is improved, but reliability deteriorates because failure of one block renders the entire metablock unusable
Solution Approach 1:
The patent implements dynamic metablock relinking where the controller periodically evaluates block health metrics and reconfigures metablock compositions. Instead of static groupings, blocks are dynamically reassigned to metablocks based on their current health status, allowing the system to adapt to block degradation over time and prevent single-point failures from compromising entire metablocks.
Solution Approach 2:
The patent segments the storage system into individual block-level units that can be independently managed and reassigned. By breaking down the traditional fixed metablock structure into flexible block assignments, the system can isolate failed blocks and redistribute their contents to healthy blocks, preventing cascade failures and improving overall reliability while maintaining write performance.
2Reliability
If data recovery and transfer is performed when a block fails, then reliability is improved, but productivity deteriorates due to resource intensive operations and extended time
Solution Approach 1:
The patent implements preliminary monitoring and evaluation of block health metrics before actual failures occur. The controller continuously assesses block conditions and proactively performs data migration and metablock reconfiguration during low-utilization periods, preventing the need for urgent recovery operations that would disrupt command response times and system productivity.
Solution Approach 2:
The patent prioritizes and accelerates recovery operations based on urgency levels. Critical data recovery operations are executed with highest priority to minimize impact on productivity, while less critical operations are deferred or performed in parallel during idle periods, ensuring that essential reliability maintenance does not significantly degrade system performance.
3Reliability
If metablocks are relinked to replace failed blocks, then reliability is improved, but loss of time occurs due to the relinking process
Solution Approach 1:
The patent implements periodic health assessments and scheduled relinking operations rather than immediate reactive relinking. By evaluating block health metrics at regular intervals and performing relinking during planned maintenance windows or low-activity periods, the system minimizes disruption to operational time while maintaining reliability through systematic updates of metablock compositions.
Data Source
AI summary
A method includes, in a data storage device that includes a non-volatile memory having multiple memory dies, determining whether one or more metablocks are metablock update candidates based on relinking metrics corresponding to the one or more metablocks. Each memory die includes multiple blocks of storage elements, and metablocks are formed through linking of blocks from the multiple memory dies. The method also includes comparing a number of the metablock update candidates to a relinking pool threshold. The method further includes, in response to the number of the metablock update candidates satisfying the relinking pool threshold, updating the linking of the blocks of the metablock update candidates to form updated metablocks. Linking of blocks may be updated by changing fields of a metablock data table, and blocks may be grouped based on block health values to extend an average useful life of the updated metablocks.