ECC Sub-block Redundancy for High-density Magnetic Disk Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional Error Correction Code (ECC) methods are insufficient for achieving high reliability in data recovery from emerging Data Storage Devices (DSDs) with higher areal densities, due to increased errors caused by smaller data scales and tighter head positioning tolerances, as well as Adjacent Track Interference and Wide Area Track Erasure issues.
Innovation Solution
The implementation of an enhanced ECC system that encodes data blocks into ECC sub-blocks and super-blocks, with redundant data for error correction, and a write-verify process that duplicates failed ECC sub-blocks in subsequent super-blocks to provide additional recovery capabilities, allowing for cascading data recovery across multiple ECC super-blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If Shingled Magnetic Recording (SMR) is used to increase Tracks Per Inch (TPI), then storage capacity increases, but Adjacent Track Interference (ATI) and Wide Area Track Erasure (WATER) worsen
Solution Approach 1:
The patent segments the recording surface into distinct track zones with different recording densities. Outer tracks use conventional recording while inner tracks use SMR, creating a segmented approach that isolates ATI-prone areas from critical data zones. This segmentation allows high-capacity SMR storage while protecting important data from interference effects.
Solution Approach 2:
The patent applies different recording quality standards to different locations on the disk. Critical data is written to outer tracks with larger margins and lower density, while less critical data occupies inner SMR tracks. This local quality differentiation ensures that data subject to ATI and WATER has enhanced protection through positioning and redundant encoding.
2Quantity of substance
If energy-assisted recording techniques (HAMR, MAMR) are used to increase areal density, then storage capacity increases, but read errors increase due to smaller data scale and tighter tolerances
Solution Approach 1:
The patent performs preliminary error detection and correction by writing verification data and known patterns alongside user data. Read-verify operations are performed immediately after writing to detect errors before they propagate. This preliminary action allows early intervention through error correction codes and data rewriting, improving reliability despite the smaller feature sizes enabled by HAMR and MAMR.
Solution Approach 2:
The patent implements feedback mechanisms where read operations verify written data and feed error information back to the write channel. When errors are detected in high-density regions, the system adjusts write parameters, rewrites affected sectors, or activates additional error correction. This closed-loop feedback compensates for the reduced noise margins inherent in high areal density storage.
3Reliability
If conventional ECC is used for error correction, then some bit errors can be corrected, but data reliability is insufficient for high areal density disks
Solution Approach 1:
The patent extends error correction from the traditional single-disk, single-track dimension to multiple dimensions including cross-track redundancy and multi-sector verification. By organizing ECC codes across multiple tracks and sectors rather than within single sectors, the system creates a multi-dimensional error correction network that can recover from errors in high-density regions where conventional single-sector ECC fails.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach significantly improves data reliability by enabling the recovery of multiple ECC sub-blocks and super-blocks, effectively handling higher error frequencies and inter-track interference, thereby maintaining data integrity in high-density DSDs.
Implementation Method 1
a head is positioned in relation to a recording surface on the disk to magnetically read and write data
Implementation Method 2
SMR increases TPI by using a shingle write head with a stronger magnetic field to overlap tracks like roof shingles
Data Source
AI summary
A Data Storage Device (DSD) includes one or more magnetic disks. One or more data blocks are encoded into a first plurality of Error Correction Code (ECC) sub-blocks including a first ECC sub-block. The first plurality of ECC sub-blocks is encoded into a first ECC super-block. The first ECC sub-block is write-verified by reading the first ECC super-block. If the write-verify passes, a second plurality of ECC sub-blocks is encoded into a subsequent ECC super-block. If the write-verify fails, the first ECC sub-block and a subset of the second plurality of ECC sub-blocks are encoded into the subsequent ECC super-block. In another aspect, in response to the first ECC super-block failing to recover the first ECC sub-block, a subsequent ECC super-block is read and a copy of the first ECC sub-block is used if the copy is detected in the subsequent ECC super-block.