Media-Aware ECC Decoding for Latency Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems lack effective error correction mechanisms, particularly in handling uncorrectable errors, which can lead to data corruption due to electrical or magnetic interference and degradation of NAND memory structures, necessitating a more robust ECC controller and operating method.
Innovation Solution
A memory system with an ECC controller that includes a processor, memory characterizer, in-flight assessor, and selective decoder, which generates an index decision table to predict and execute error recovery actions, employing hard decoding, read retry decoding, OVS two-bit decoding, and soft decoding in a sequential or adaptive manner to correct errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional ECC decoding is used without media quality awareness, then the system structure is simple, but error correction reliability is insufficient and data corruption occurs
Solution Approach 1:
The memory characterizer performs preliminary characterization of media quality attributes (program-erase cycle count, read-disturb count, retention time) before actual read operations. This advance assessment allows the system to predict error rates and select appropriate decoding strategies in advance, improving reliability without adding complex real-time decision-making structures.
Solution Approach 2:
The ECC controller dynamically adapts its decoding strategy based on predicted media quality. Instead of using a fixed decoding approach, the system adjusts the selection of decoding algorithms (hard decoding, soft decoding, read-retry) according to the characterized media state, optimizing error correction effectiveness for different media conditions.
2Reliability
If multiple decoding algorithms are always executed to ensure error correction, then error correction capability is maximized, but data access latency increases
Solution Approach 1:
Based on media quality prediction, the system executes only the necessary portion of the decoding algorithm sequence. For high-quality media, minimal decoding (hard decoding only) is performed. For degraded media, additional decoding passes (soft decoding, read-retry) are selectively activated. This partial execution approach maintains error correction capability while minimizing unnecessary processing time.
Solution Approach 2:
The in-flight assessor uses the index decision table to skip unnecessary decoding algorithms based on predicted media quality. When media is characterized as high-quality, the system rushes through with only essential decoding steps, bypassing redundant error correction attempts that would waste time. This selective skipping reduces latency while maintaining adequate error correction.
3Measurement precision
If comprehensive media characterization is performed, then decoding accuracy is improved, but processing overhead and power consumption increase
Solution Approach 1:
The memory characterizer extracts only the most critical media quality attributes (program-erase cycle count, read-disturb count, retention time) that have the greatest impact on error rates. By focusing on these key parameters rather than comprehensively analyzing all possible media characteristics, the system achieves sufficient assessment accuracy with reduced processing overhead and lower power consumption.
Solution Approach 2:
The memory system performs self-characterization by monitoring its own operational history and media state. The controller tracks program-erase cycles, read-disturb events, and retention times internally without requiring external characterization tools. This self-service approach provides accurate media quality assessment while minimizing additional processing power requirements.
4Reliability
If uncorrectable errors are not handled properly, then the system operates quickly, but data corruption and reliability issues occur
Solution Approach 1:
The system takes preliminary anti-action by predicting when uncorrectable errors are likely to occur based on media quality characterization. Before actual errors manifest, the system prepares appropriate error recovery strategies and selects robust decoding algorithms in advance. This proactive approach prevents data corruption while maintaining operational speed by avoiding reactive error handling delays.
Data Source
AI summary
A memory system and operating method thereof includes a semiconductor memory device, and a memory controller controlling actions of the memory device. The memory controller contains a processor executing instruction and programs stored in the memory controller, a memory characterizer characterizing the memory system, and generating an index decision table, an in-flight assessor assessing read command, and predicting a proposed error recovery action in accordance with the index decision table, and a selective decoder executing the proposed error recovery action.