SSD Defect Map Persistence for Sub-Block Read Fail Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional firmware in SSDs retires entire blocks upon error correction code (ECC) fails, impacting performance, write amplification, and reducing overprovisioning, while maintaining read fail tables in RAM increases complexity and susceptibility to corruption.
Innovation Solution
Implementing a firmware that updates a read defect map through dedicated opcodes and an OTP page to track word line and sub-block group pages, retiring only affected pages during read fails, and maintaining the table in NAND media to reduce the need for RAM-based table structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If entire blocks are retired upon ECC fails, then data reliability is improved, but performance deteriorates and overprovisioning is reduced
Solution Approach 1:
The patent segments the retirement process from block level to sub-block level. Instead of retiring entire blocks when ECC fails occur, only the specific sub-blocks containing failed pages are retired. This segmentation allows the majority of the block to remain usable, thereby maintaining performance and overprovisioning while still ensuring data reliability through targeted defect management.
2Adaptability or versatility
If complex tables are maintained in RAM to track defects, then defect tracking capability is improved, but system reliability deteriorates due to corruption risk during power interruptions
Solution Approach 1:
The patent extracts the defect map data structure from volatile RAM and stores it directly in the OTP area of the NAND media. This extraction eliminates the risk of corruption during power interruptions while maintaining full defect tracking capability. The OTP area provides non-volatile storage that preserves defect information across power cycles without requiring complex RAM-based table structures.
Solution Approach 2:
The system uses the OTP area of the NAND media itself to store the defect map, making the storage medium self-sufficient for tracking its own defects. This self-service approach eliminates the need for external RAM tables and ensures that defect information is always available and reliable, as it is stored in the same non-volatile medium that requires the defect tracking.
3Measurement precision
If RAM tables are used to track sub-block defects, then defect tracking precision is improved, but device complexity increases
Solution Approach 1:
The patent merges the defect map storage with the NAND media's OTP area, combining the tracking function directly into the storage medium. This merging eliminates the need for separate RAM tables and complex firmware management logic, while maintaining sub-block level defect tracking precision. The defect map becomes an integral part of the NAND media structure rather than a separate computational component.
Data Source
AI summary
An embodiment of an electronic apparatus may include one or more substrates, and logic coupled to the one or more substrates, the logic to control access to a persistent storage media based on a block and sub-block access structure, store a data structure in the persistent storage media to track read fails at a sub-block granularity for a word-line for every block, and update the data structure in response to a read fail on a block to indicate a failed sub-block that corresponds to the read fail for a word-line for the block. Other embodiments are disclosed and claimed.