Flash Memory Encoding History Metadata for Page Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Flash memory devices, particularly NAND flash devices, lack random access capabilities, making it difficult to improve data protection and security, as they require sequential commands for accessing data, which complicates the recovery of damaged pages.
Innovation Solution
A method is introduced to access encoding-history information by creating a super-block structure across multiple flash units, where encoding-history information is programmed into a metadata section of a designated page, allowing for the recovery of damaged pages by identifying unused or unencoded pages within the super-page strings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If NAND flash devices use sequential access method to write commands and addresses, then the device can operate reliably, but random access capability is lost making data protection and damaged page recovery difficult
Solution Approach 1:
The patent applies preliminary action by pre-programming encoding-history information into the metadata section of designated pages during normal operation. This historical encoding data is prepared in advance so that when page damage occurs, the recovery process can directly utilize this pre-stored information without needing complex sequential access to reconstruct encoding data, thus improving both reliability and operational ease.
Solution Approach 2:
The patent introduces encoding-history information as an intermediary element stored in the metadata section. This intermediary contains historical encoding data that mediates between the damaged page and the recovery process, enabling direct recovery operations without requiring complex sequential access patterns, thereby resolving the contradiction between reliability and ease of operation.
2Reliability
If encoding-history information is stored in metadata section of each page, then damaged page recovery is enabled, but storage space is consumed for history entries
Solution Approach 1:
The patent applies local quality by selectively storing encoding-history information only in the metadata section of designated pages rather than all pages. The history entries are locally placed where they are most needed for recovery operations, and the metadata section is utilized efficiently to store this information without consuming additional user data space, thus balancing reliability improvement with storage space conservation.
3Reliability
If super-block structure is created across multiple flash units, then data protection is enhanced through distributed storage, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the storage system into super-blocks that span multiple flash units, with each super-block containing designated pages that have encoding-history information. This segmentation allows distributed storage of protection information across multiple physical units while maintaining manageable complexity through structured organization of super-pages and metadata sections.
Data Source
AI summary
The invention relates to a method, a non-transitory computer-readable storage medium and an apparatus for accessing to encoding-history information. The method includes: providing a super-block formed by storage space of flash units, where the super-block includes zones, each zone includes super-page strings, and each super-page string includes pages across the flash units; and programming encoding-history information into a metadata section of a designated first page of a designated super-page string, thereby enabling a damaged page that is occurred in the designated super-page string of the designated zone to be recovered according to the encoding-history information. The encoding-history information includes a history profile and history entries. The history profile indicates which zone or zones are covered in the super-block, and a quantity of the history entries. Each history entry includes information indicating that a designated second page of the designated super-page string in a designated zone hasn't been passed through an engine to generate a parity for the designated super-page string.