Error Correction Decoding Using Threshold-Triggered Redundancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current error correction methods in data storage systems, particularly in storage-class memory (SCM), face challenges in balancing reliability, cost efficiency, and latency due to the limitations of existing error correction codes, which can lead to undetected errors and increased storage overhead.
Innovation Solution
An enhanced error correction system that employs a threshold limit below the error correction limit to initiate a secondary correction process using redundancy data, reducing undetected errors and storage overhead by selectively applying error correction schemes and redundancy mechanisms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If more error correction overhead is added to payload data during encoding, then data reliability is improved, but storage space consumption increases and memory cost increases
Solution Approach 1:
The patent applies partial error correction by setting a threshold limit below the full error correction limit. The error correction circuit performs corrections only up to this threshold, and only when the number of detected errors exceeds this threshold does the system retrieve redundant data. This partial correction approach reduces storage overhead compared to full error correction while maintaining acceptable reliability levels.
2Reliability
If more error correction overhead is added to payload data during encoding, then data reliability is improved, but decoding complexity increases
Solution Approach 1:
The patent segments the error correction process into two distinct stages: (1) initial error correction using the error correction circuit with a threshold limit, and (2) conditional retrieval of redundant data using the detection circuit when errors exceed the threshold. This segmentation allows the system to handle most cases with simple threshold comparison rather than complex full decoding, reducing overall decoding complexity while maintaining reliability.
Solution Approach 2:
The system performs partial error correction only when necessary (when errors exceed the threshold), rather than always performing full error correction. This conditional approach reduces the average decoding complexity by avoiding unnecessary complex decoding operations for data with few or no errors.
3Reliability
If error correction threshold limit is set below error correction limit, then undetected errors are reduced, but additional redundancy data storage is required
Solution Approach 1:
The patent changes the error correction threshold parameter from the full error correction limit to a lower threshold value. This parameter change enables the detection circuit to identify cases where errors are present but below the full correction capability, allowing conditional retrieval of redundant data. This approach reduces undetected errors while the redundancy storage is only activated when needed, optimizing the trade-off between reliability and storage overhead.
4Reliability
If conventional error correction codes are used, then data storage reliability is improved, but latency increases due to media and latency limitations
Solution Approach 1:
The system performs partial error correction with a threshold limit rather than always performing full error correction. This reduces the average processing time and latency by handling most data with simple threshold comparison and only invoking full redundant data retrieval when necessary, thus maintaining reliability while reducing latency penalties.
Solution Approach 2:
The error correction circuit performs preliminary error correction and threshold checking before determining whether full redundant data retrieval is necessary. This preliminary action filters out most cases that don't require complex correction, reducing overall latency by preparing the data in advance and only engaging resource-intensive operations when truly needed.
Data Source
AI summary
Enhanced error correction for data stored in storage devices are presented herein. An error correction circuit decodes an encoded data segment retrieved from a storage media. This decode uses a selected error correction scheme having an error correction limit. The error correction circuit tracks a number of bit corrections made to the encoded data segment during decode. A detection circuit sends a redundant version of the encoded data segment to the error correction circuit in response to the number of bit corrections satisfying a threshold limit set below the error correction limit to mitigate undetected errors in decoding the encoded data segment. An output circuit can transfer resultant data decoded by the error correction circuit to other systems, such as a host device.