Partial Cloning of Compressed Data Containers Without Full Recompression

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing storage systems face inefficiencies in partial cloning with compression, as they use non-variable compression group sizes, which require decompressing and re-compressing all data blocks, making them inefficient for partial data container cloning.

Innovation Solution

Implementing a method and system that determines and replicates compressed blocks using a non-variable compression group size while decompressing and storing blocks not within this size, allowing for partial cloning with improved efficiency by using both variable and non-variable compression group sizes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If non-variable compression group size is used for compressing data blocks, then compression efficiency is improved for full data container cloning, but partial cloning becomes inefficient due to requiring decompression and re-compression of all blocks

Engineering Contradiction:
Improvecompression efficiencyVSAvoidpartial cloning efficiency
Core Design Contradiction:
ProductivityVSEase of operation

Solution Approach 1:

The data container is segmented into multiple data extents, where each extent can be independently cloned. This segmentation allows partial cloning operations to work on specific extents without affecting others, enabling efficient partial cloning while maintaining compression benefits for each segment individually

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different compression approaches are applied to different parts of the data structure. Compressed blocks within a data extent maintain non-variable compression group size for efficiency, while the extent-level segmentation enables selective partial cloning. This local differentiation resolves the contradiction between full compression efficiency and partial cloning flexibility

Inventive Principle:
Principle #3Local quality

2Manufacturing precision

If all data blocks are decompressed for partial cloning, then accurate cloning is achieved, but processing time and computational resources increase significantly

Engineering Contradiction:
Improvecloning accuracyVSAvoidprocessing time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

Data blocks are pre-compressed using non-variable compression group sizes before storage. This preliminary compression action allows the system to store compressed data efficiently while maintaining the ability to perform partial cloning by working with extent-level structures, avoiding the need to decompress individual blocks during cloning operations

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system creates copies at the data extent level rather than at the individual block level. This copying approach maintains cloning accuracy by preserving the compressed block structures within extents, while significantly reducing processing time by avoiding block-level decompression and re-compression operations

Inventive Principle:
Principle #26Copying

Data Source

PatentUS8866649B2Method and system for using non-variable compression group size in partial cloning
Publication Date: 2014.10.21 NETAPP INC
  • US8866649B2 patent drawing
  • US8866649B2 patent drawing
  • US8866649B2 patent drawing

AI summary

Method and system for partially cloning a data container with compression is provided. A storage operating system determines if a portion of a source data container that is to be cloned includes a plurality of compressed blocks that are compressed using a non-variable compression group size. The operating system clones the plurality compressed blocks with the non-variable compression group size and de-compresses a plurality of blocks of the data container that are not within the non-variable compression group size. The plurality of compressed blocks and the plurality of blocks that are not within the non-variable compression group size are then stored as a partially cloned copy of the source data container.