Multi-Layer ECC Parity for SMR Write Throughput and Data Recovery

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Shingled magnetic recording (SMR) hard disk drives face performance degradation due to the write verify function, which is necessary for data reliability but decreases write command throughput, and existing error correction coding (ECC) schemes do not adequately address data integrity and recovery in high-capacity digital storage systems.

Innovation Solution

A multi-layer error correction coding (ECC) parity technique is implemented, where data bands are divided into sub-data bands, generating multiple layers of parity matrices to enable efficient recovery of corrupt data, allowing the system controller to determine and retrieve necessary parity information for error correction, thereby reducing the need for write verify operations and enhancing data integrity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If write verify function is used to ensure data reliability, then data integrity is improved, but write command throughput deteriorates

Engineering Contradiction:
Improvedata integrityVSAvoidwrite command throughput
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies preliminary action by generating multi-layer ECC parity information during the write operation before actual data storage. This pre-computed parity information is stored alongside the data, enabling error correction to occur during subsequent read operations without requiring additional write verify cycles. The system performs error correction preparation in advance, eliminating the need for time-consuming post-write verification.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates multiple copies of parity information across different layers (first-layer parity for sub-data bands, second-layer parity for data bands, and optional third-layer parity for radial zones). These redundant parity copies enable the system to recover corrupt data through read operations alone, eliminating the need for write verify functions that would otherwise be required to ensure data integrity.

Inventive Principle:
Principle #26Copying

2Reliability

If multi-layer ECC parity technique is implemented for efficient error correction, then data recovery capability is improved, but system complexity increases

Engineering Contradiction:
Improvedata recovery capabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the error correction system into multiple hierarchical layers: first-layer ECC parity divided by sub-data bands, second-layer ECC parity divided by data bands, and optional third-layer parity divided by radial zones. Each layer operates independently with its own parity generation and correction logic, allowing the system to handle different error scenarios at appropriate granularity levels without requiring a monolithic complex correction system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extends error correction from traditional two-dimensional track/sector organization to three dimensions by adding radial zone-based third-layer parity. This dimensional expansion allows the system to correct errors that span multiple tracks and zones simultaneously, improving data recovery capability while organizing complexity across multiple spatial dimensions rather than increasing operational complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Quantity of substance

If shingled magnetic recording is used to increase areal density, then storage capacity is improved, but data track update capability deteriorates

Engineering Contradiction:
Improvestorage capacityVSAvoiddata track update capability
Core Design Contradiction:
Quantity of substanceVSEase of operation

Solution Approach 1:

The patent applies preliminary action by pre-calculating and storing multi-layer ECC parity information that spans across overlapping shingled tracks. When a track update is required, the system can retrieve the necessary parity information from adjacent tracks and the multi-layer parity structures, perform corrections, and reconstruct updated data without requiring physical track erasure or complex sequential rewriting operations.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces multi-layer ECC parity information as an intermediary that mediates between corrupt data in shingled tracks and the recovery process. This intermediary parity data, distributed across multiple layers and tracks, enables the system to resolve track update conflicts and recover corrupted information without directly manipulating the overlapping track structure, thereby simplifying update operations in SMR systems.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS10275309B2Multi-layer integrated zone partition system error correction
Publication Date: 2019.04.30 SANDISK TECHNOLOGIES LLC
  • US10275309B2 patent drawing
  • US10275309B2 patent drawing
  • US10275309B2 patent drawing

AI summary

A multi-layer error correction coding (ECC) parity technique involves dividing a data band into sub-data bands, generating a respective 1.sup.st-layer sub-data band parity matrix for each associated sub-data band, and generating a respective (q.sup.th>1)-layer parity matrix for sets of associated adjacent sub-data bands. In the context of a data storage system, the parity generation may be performed at the system-side, and communicated and written to one or more associated data storage devices (DSDs) along with the corresponding data, whereby the DSDs may further associate track ECC information to the written data. In response to receiving at the system-side, location-identifying information about data errors that are not correctable by the DSD using track ECC information, the system may determine an amount of the multi-layer parity information needed to recover the corrupt data, and make a data/parity read request accordingly.