Exception Regions for Shingled Disk Drive Stream Handling

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current hard disk drives, particularly Shingled Disk Drives using shingled perpendicular magnetic recording, face challenges in efficiently handling fragmented data streams and sustaining consecutive write operations under varying conditions, leading to data integrity issues and degradation over time.

Innovation Solution

The implementation of Exception Regions in disk drives, managed in states of free, active, recovery, and used, with TRIM commands to indicate deleted content and associate random write commands for efficient defragmentation and data management, allowing for enhanced fragmented stream handling without significant changes to the SMR architecture.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If shingled perpendicular magnetic recording (SMR) is used to achieve high track density, then storage capacity is improved, but data integrity and fragmentation handling deteriorate

Engineering Contradiction:

Improvestorage capacityVSAvoiddata integrity

Core Design Contradiction:

Quantity of substanceVSReliability

Solution Approach 1:

The disk drive is segmented into multiple zones with different recording densities, allowing the system to optimize for both capacity and reliability in different regions. Exception regions are created within the disk structure to handle fragmented data separately from the main SMR storage areas, enabling independent management of data integrity concerns.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Exception regions act as intermediary zones between the high-density SMR storage areas and the data management system. These regions buffer and manage fragmented data streams, preventing fragmentation issues from propagating to the main storage areas while maintaining the high capacity benefits of SMR.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If multiple consecutive write streams are sustained, then productivity is improved, but data fragmentation increases

Engineering Contradiction:

Improvewrite stream throughputVSAvoiddata fragmentation

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

Exception regions are pre-configured and maintained in a ready state before fragmentation occurs. When multiple write streams are initiated, the system can immediately route fragmented data to these pre-prepared exception regions, preventing fragmentation from degrading performance while sustaining high write throughput.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system discards fragmented data into exception regions for temporary isolation, then recovers it through defragmentation operations. This allows continuous write operations to proceed at high speed while fragmentation is handled separately, maintaining both productivity and data organization.

Inventive Principle:

Principle #34Discarding and recovering

3Reliability

If exception regions are implemented for fragmented stream handling, then data integrity is improved, but device complexity increases

Engineering Contradiction:

Improvedata integrityVSAvoidregion management complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

Exception regions are designed to serve multiple functions: handling fragmented data, buffering write streams, and providing a recovery mechanism for data integrity. This multi-functionality reduces the need for separate specialized components, thereby limiting the increase in device complexity while achieving improved data integrity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Stability of the object's composition

If defragmentation operations are performed on used regions, then data organization is improved, but loss of time occurs

Engineering Contradiction:

Improvedata organizationVSAvoiddefragmentation time

Core Design Contradiction:

Stability of the object's compositionVSLoss of time

Solution Approach 1:

Defragmentation of exception regions is performed periodically rather than continuously or on-demand. This periodic approach allows the system to maintain data organization in exception regions without causing continuous interruptions to write operations, thereby balancing data organization benefits with minimal time loss.

Inventive Principle:

Principle #19Periodic action

Data Source

PatentUS8654472B2Implementing enhanced fragmented stream handling in a shingled disk drive

Publication Date: 2014.02.18 WESTERN DIGITAL TECHNOLOGIES INC

AI summary

A method and a storage system are provided for implementing fragmented stream handling for persistent media including shingled disk drives (SDDs). A plurality of Exception Regions for recording stream commands is defined for disk regions. One state is maintained in each of the Exception Regions including a free Exception Region, an active Exception Region, a recovery Exception Region and a used Exception Region. TRIM commands are issued by a host to the drive to indicate deleted content.