Data Recovery Without Sync Mark Retry in Storage Read Channels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data processing systems, such as storage systems, face issues with data corruption due to missing sync marks in data fragments, leading to synchronization failures and increased access times, resulting in potential data loss.
Innovation Solution
The implementation of a no sync mark retry (NSMR) technique in a data processing apparatus, where a sync mark is assumed at a selected position in a data fragment missing a sync mark, and the data is processed to determine convergence, allowing for efficient recovery of the data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the sync mark detector waits for the disk to complete a revolution and retry detection, then the likelihood of detecting the sync mark improves, but the overall access time increases
Solution Approach 1:
The patent applies preliminary action by performing data processing and convergence determination before waiting for the next disk revolution. The system processes the data fragment assuming a sync mark position, checks for convergence, and only then waits for retry if needed. This preliminary processing reduces the overall access time by utilizing the waiting period more effectively.
Solution Approach 2:
The patent maintains continuity of useful action by continuously processing data fragments through convergence determination even when sync mark detection fails. Instead of idle waiting, the system continuously attempts data processing with assumed sync mark positions, ensuring that when the disk completes a revolution, the system is ready to immediately validate or retry without additional delay.
2Measurement precision
If the sync mark detector requires accurate sync mark detection to enable channel decoder operation, then data accuracy improves, but data loss occurs when sync mark detection fails
Solution Approach 1:
The patent introduces an intermediary convergence determination process between sync mark detection and channel decoder operation. Instead of directly enabling the channel decoder based solely on sync mark detection, the system first processes the data fragment assuming a sync mark position and determines convergence. This intermediary step validates whether the assumed sync mark position is correct before enabling channel decoding, preventing data loss from incorrect assumptions.
Solution Approach 2:
The system applies self-service by using the data fragment itself to determine the correct sync mark position through convergence determination. The data processing reveals whether the assumed sync mark position is correct based on whether the processed data converges to valid results, allowing the system to self-validate without external intervention or additional hardware.
3Reliability
If the read channel initiates a retry when sync mark detection fails, then the reliability of data recovery improves, but the productivity of the storage system decreases
Solution Approach 1:
The patent applies dynamics by making the retry process adaptive rather than static. The system dynamically determines whether to retry based on convergence determination results. If data processing converges successfully with an assumed sync mark position, the system accepts the data without waiting for a full disk revolution retry. Only when convergence fails does the system initiate the traditional retry process, thereby reducing unnecessary waits and improving productivity while maintaining reliability.
Solution Approach 2:
The patent changes the operational parameter from binary (detect sync mark or retry) to a spectrum including convergence determination. The system processes data with assumed parameters (sync mark position) and changes the retry decision based on whether processing converges. This parameter change allows the system to accept data on first attempt when convergence succeeds, improving throughput while maintaining data recovery reliability through the convergence check.
Data Source
AI summary
A read channel is configured to receive at least part of a data fragment read from a storage media into a register, wherein the data fragment is configured to be formatted with a preamble, a sync mark (e.g., a syncMark), and user data, and wherein the data fragment is missing a sync mark. A position in the data fragment is selected, a sync mark is assumed at the selected position. The data is then processed assuming the sync mark is at the selected position of the data fragment to determine whether the data converges. When a determination is made that the data converges, the data is recovered.