Zone Provisioning and Compression for Disk Drives

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Traditional storage provisioning methods lead to performance degradation as physical space fills, and they lack flexibility in managing varying zone sizes, which affects data storage efficiency and performance.

Innovation Solution

Implementing thin provisioning with zone-level granularity and data compression, treating storage devices as Shingled Magnetic Recording (SMR) drives, and dynamically managing physical space to create logical zones larger than physical zones, allowing for efficient allocation and reuse of storage space.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If traditional storage provisioning methods are used to allocate physical space, then initial storage capacity is provided, but performance degrades as physical space fills

Engineering Contradiction:

Improvestorage capacityVSAvoidperformance stability

Core Design Contradiction:

Quantity of substanceVSReliability

Solution Approach 1:

The storage device is divided into multiple zones with different provisioning characteristics. Inner zones use traditional provisioning while outer zones use thin provisioning with compression, allowing the system to segment storage management to avoid performance degradation in filled regions and maintain overall system reliability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The provisioning method dynamically adapts based on zone fill status. When zones become full, the system can remap logical addresses to different physical zones, dynamically adjusting the provisioning strategy to maintain performance while continuing to provide storage capacity.

Inventive Principle:

Principle #15Dynamics

2Adaptability or versatility

If fixed zone sizes are used in traditional provisioning, then simple management is achieved, but flexibility in managing varying zone sizes is lost

Engineering Contradiction:

Improvezone size flexibilityVSAvoidprovisioning management complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

Storage is segmented into zones with different sizes and provisioning characteristics. Each zone can be independently configured with appropriate granularity and compression settings, providing flexibility to match varying workload requirements while keeping individual zone management relatively simple.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Different zones are assigned different provisioning qualities - inner zones may use traditional methods while outer zones use thin provisioning with compression. This local differentiation allows optimal performance for different access patterns without requiring complex uniform management across the entire device.

Inventive Principle:

Principle #3Local quality

3Quantity of substance

If logical address spaces are made larger than physical space to enable thin provisioning, then storage efficiency improves, but address management complexity increases

Engineering Contradiction:

Improvestorage efficiencyVSAvoidaddress space management

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

A translation layer acts as an intermediary between logical addresses and physical zones. This mediator handles the complexity of mapping sparse logical addresses to physical storage, managing remapping when zones fill, and coordinating compression operations without exposing this complexity to users.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system maintains mapping tables that copy and track logical-to-physical address relationships. These copies enable efficient lookup and remapping operations, allowing thin provisioning with large logical address spaces while managing complexity through pre-computed mappings rather than real-time calculations.

Inventive Principle:

Principle #26Copying

4Quantity of substance

If compression is applied to increase storage capacity, then data density improves, but processing overhead increases

Engineering Contradiction:

Improvedata densityVSAvoidprocessing overhead

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

Compression is applied selectively to specific zones rather than uniformly across all storage. This segmentation allows compression to be enabled in zones where it provides benefit while avoiding unnecessary processing overhead in zones where data is already densely packed or access patterns make compression less valuable.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The compression system operates autonomously within provisioned zones, automatically managing compression ratios, selecting compression algorithms, and handling decompression during reads without requiring external intervention or complex coordination, thereby reducing overall system processing overhead.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS10795597B2Thinly provisioned disk drives with zone provisioning and compression in relation to zone granularity

Publication Date: 2020.10.06 SEAGATE TECH LLC

AI summary

Methods, systems, and apparatuses are described for provisioning storage devices. An example method includes specifying a logical zone granularity for logical space associated with a disk drive. The method further includes provisioning a zone of a physical space of the disk drive based at least in part on the specified logical zone granularity. The method also includes storing compressed data in the zone in accordance with the provisioning.