SMR Track Erasure Coding for Recovering Squeezed Sectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Shingled Magnetic Recording (SMR) systems face challenges in recovering 'squeezed' sectors, where arbitrary numbers of erased sectors within a track need to be replaced without entering Data Recovery Procedures (DRP), and existing error correction codes are not scalable or efficient for SMR architectures.
Innovation Solution
The implementation of a programmable Cauchy-type track erasure correction code combined with a media-error correction code generates N-weighted parity sectors per track, allowing for the replacement of up to N erased sectors per track, using a Track Erasure Decoder (TED) that ensures data integrity through a 'data-integrity handshake' and on-the-fly hardware decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If conventional sector-based ECC is used in SMR drives, then implementation is simple, but it cannot efficiently handle arbitrary sector replacements in shingled tracks
Solution Approach 1:
The patent divides the track into multiple segments or groups of sectors, with each segment having its own dedicated repair block. This segmentation allows the ECC system to handle arbitrary sector replacements within each segment independently, while maintaining overall track integrity. The repair blocks are organized in a hierarchical structure that enables flexible recovery of squeezed sectors without requiring complete track reprocessing.
Solution Approach 2:
The patent introduces an additional dimension to the traditional ECC structure by organizing repair blocks not just per sector but per group of sectors across multiple tracks. This multi-dimensional organization allows the system to handle arbitrary sector replacements by selecting appropriate repair blocks from different dimensional layers, providing versatility while maintaining manageable complexity.
2Reliability
If large ECC block sizes are used to improve error correction capability, then more erased sectors can be corrected, but processing latency and complexity increase
Solution Approach 1:
The patent segments the large ECC block into smaller sub-blocks or groups, each with its own repair block. This allows the error correction process to be performed on smaller units independently and in parallel, reducing processing latency while maintaining the ability to correct a large total number of erased sectors across the entire track through the hierarchical repair structure.
Solution Approach 2:
The patent pre-calculates and stores repair blocks for each segment before they are needed for recovery. These repair blocks are generated in advance and stored in the hierarchical repair structure, allowing immediate recovery of erased sectors without performing complex real-time calculations, thus reducing processing latency while maintaining strong error correction capability.
3Adaptability or versatility
If multiple repair blocks are used to handle arbitrary sector replacements, then more erased sectors can be recovered, but system complexity increases
Solution Approach 1:
The patent organizes multiple repair blocks into a hierarchical structure with clear segmentation at different levels (sector-level, track-level, and global-level repair blocks). This structured segmentation allows the system to handle arbitrary sector replacements systematically, where each level of the hierarchy addresses specific recovery scenarios, reducing the apparent complexity by providing a organized framework for managing multiple repair blocks.
Solution Approach 2:
The patent designs the repair blocks to serve multiple functions across different levels of the hierarchy. Each repair block can be used for its specific segment's recovery or combined with other repair blocks for broader recovery scenarios. This multi-functionality reduces the total number of repair blocks needed compared to having dedicated repair blocks for every possible failure scenario, thereby reducing system complexity while maintaining high adaptability.
Data Source
AI summary
A technique for recovering of “squeezed” sectors in a set of sequential sectors such as are used in Shingled Magnetic Recording (SMR) is described. Embodiments of the invention use a programmable erased sector recovery scheme, which is a concatenation of a “Cauchy-type” track erasure correction code, together with a media-error correction code that generates N-weighted parity-sectors per track and is capable of replacing up to N-erased sectors per track in any possible combination.