NAND Flash Write Processing for Unchanged Data Metadata
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current techniques for modifying data in NAND Flash storage require a significant number of accesses and processing power, even for small changes, due to its characteristics of erasing data before reprogramming, which is inefficient in terms of power and time.
Innovation Solution
A storage system that performs a first write process for modifying user data and a second write process for unchanged data, where the second process bypasses the error correction encoder to reduce processing cycles and power consumption by generating new metadata and combining error correction parity information without reprocessing unchanged user data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If current NAND Flash write techniques are used to modify data, then data can be stored and modified, but a significant number of accesses and processing cycles are required even for small changes
Solution Approach 1:
The write operation is segmented into two distinct processes: a first write process for modifying user data and a second write process for unchanged data. This segmentation allows the system to treat different data scenarios differently, avoiding unnecessary processing for unchanged data while maintaining full functionality for modified data, thereby reducing overall processing time and improving write operation efficiency
Solution Approach 2:
The error correction encoding process is extracted and separated from the standard write process. The patent introduces a selective error correction process that only applies to modified data portions, rather than encoding all data. This extraction reduces the processing burden for unchanged data while maintaining data integrity for modified data, directly addressing the time loss issue
2Productivity
If current NAND Flash write techniques are used to modify data, then data can be stored and modified, but significant power consumption is required even for small changes
Solution Approach 1:
The write operation is segmented into two distinct processes: a first write process for modifying user data and a second write process for unchanged data. This segmentation allows the system to treat different data scenarios differently, avoiding unnecessary processing for unchanged data while maintaining full functionality for modified data, thereby reducing overall processing time and improving write operation efficiency
Solution Approach 2:
Instead of performing complete error correction encoding for all data (excessive action), the patent applies error correction only to the necessary portions (partial action). The selective error correction process encodes only modified data or combines old and new parity information only when needed, reducing energy consumption while maintaining sufficient data integrity protection
3Reliability
If data is rewritten in NAND Flash, then data can be updated, but the erase-then-program characteristic causes significant processing overhead
Solution Approach 1:
The patent performs preliminary actions by maintaining old parity information and user data in memory during the write process. Instead of immediately erasing and reprogramming, the system prepares both old and new data structures, then combines them selectively. This preliminary preparation simplifies the actual write operation and reduces processing complexity while ensuring data integrity through the combination of old and new parity information
Data Source
AI summary
Old user data, old metadata, and old error correction parity information are received. New metadata corresponding to the old user data is generated. The old metadata and the new metadata are combined to obtain combined metadata. New error correction parity information is generated using the combined metadata. The old error correction parity information and new error correction parity information are combined to obtain combined error correction parity information. The old user data, new metadata, and combined error correction parity information are stored in solid state storage.