Discrete Back End Allocation Units for Inline Compression

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current data storage systems face inefficiencies in managing data across multiple storage arrays and host applications, particularly in optimizing storage allocation units to balance compressibility and resource utilization, leading to suboptimal performance and space utilization.

Innovation Solution

The system organizes tangible data storage devices into discrete size back end allocation units, allowing computing nodes to select the appropriate unit based on data compressibility, available space, and device management, maintaining a 1-to-1 relationship with fixed-size front end allocation units while enabling flexible storage of both uncompressed and compressed data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If fixed-size front end allocation units are used for production volumes, then simplicity in host application interface is maintained, but storage space utilization deteriorates when data is compressed

Engineering Contradiction:

Improvehost application interface simplicityVSAvoidstorage space utilization

Core Design Contradiction:

Ease of operationVSQuantity of substance

Solution Approach 1:

The patent segments the storage allocation into two distinct layers: front end allocation units with fixed sizes that present a uniform interface to host applications, and back end allocation units with variable sizes that optimize actual storage utilization. This segmentation allows the system to maintain interface simplicity while achieving efficient space utilization through compression.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by allowing different allocation units at the back end to have different sizes based on the actual data characteristics and compression ratios of specific front end allocation units. Each back end allocation unit is tailored to the local data requirements, optimizing storage efficiency while maintaining uniform front end allocation.

Inventive Principle:

Principle #3Local quality

2Quantity of substance

If data is compressed to improve storage efficiency, then storage space utilization improves, but data access time may deteriorate due to decompression overhead

Engineering Contradiction:

Improvestorage space utilizationVSAvoiddata access time

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The patent performs data compression in advance during the data writing phase, so that when data is read, the compressed data is already in storage and can be quickly decompressed or directly accessed. This preliminary compression action ensures that storage efficiency is optimized without significantly impacting access time, as the compression work is done beforehand rather than in real-time during data retrieval.

Inventive Principle:

Principle #10Preliminary action

3Quantity of substance

If variable size back end allocation units are implemented to optimize compression, then storage efficiency improves, but system complexity deteriorates

Engineering Contradiction:

Improvestorage efficiencyVSAvoidallocation unit management complexity

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary layer (the back end allocation units with variable sizes) between the fixed front end allocation units and the physical storage devices. This intermediary handles the complexity of variable-size management, compression, and optimization, while presenting a simple fixed-size interface to host applications. The intermediary absorbs the system complexity away from the host interface.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The storage system automatically manages the variable size allocation units through self-service mechanisms, where the system itself performs compression, decompression, and allocation optimization without requiring complex external management or intervention. This automation reduces the operational complexity despite the variable size structure.

Inventive Principle:

Principle #25Self-service

4Speed

If uncompressed data is stored to maintain fast access, then data access speed improves, but storage space utilization deteriorates

Engineering Contradiction:

Improvedata access speedVSAvoidstorage space utilization

Core Design Contradiction:

SpeedVSQuantity of substance

Solution Approach 1:

The patent implements dynamic allocation where the compression and storage format of data is not fixed but adapts based on data characteristics, access patterns, and available storage space. The system can dynamically choose to store data in compressed or uncompressed form, and can dynamically adjust the size of back end allocation units based on actual needs, optimizing both access speed and space utilization in a flexible manner.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS10496278B1Inline compression support using discrete sized containers for backing store

Publication Date: 2019.12.03 EMC IP HLDG CO LLC

AI summary

A storage array presents a logical production volume that is backed by tangible data storage devices. The production volume is organized into fixed size front end allocation units. The tangible data storage devices are organized into discrete size back end allocation units of a plurality of different sizes. Data associated with each one of the front end allocation units is stored on only one of the back end allocation units. For example, compressed data may be stored on a back end allocation unit that is smaller than a front end allocation unit while maintaining a 1-to-1 relationship between the front end allocation unit and the back end allocation unit.