Storage Data Compression Selection for Fast Decompression

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing data compression methods in storage systems face challenges in accurately predicting decompression time for data with varying attributes, leading to adverse impacts on read performance, especially when selecting compression methods based on past decompression history.

Innovation Solution

A data compression system that estimates decompression time by counting the number of appearances of code categories in compressed data and selects appropriate compression algorithms based on a threshold value to ensure efficient decompression performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of substance

If multiple compression algorithms are used to improve compression ratio, then data reduction rate is improved, but decompression time increases and read performance deteriorates

Engineering Contradiction:

Improvedata reduction rateVSAvoiddecompression time

Core Design Contradiction:

Loss of substanceVSLoss of time

Solution Approach 1:

The patent changes the parameter of decompression time estimation by counting code categories in compressed data. This allows the system to evaluate whether a compression algorithm is suitable based on its decompression time characteristics, enabling selection of algorithms that maintain both good compression ratios and acceptable decompression speeds.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent dynamically selects compression algorithms based on real-time analysis of compressed data characteristics (code category counts). Instead of using a fixed compression method, the system adapts its choice of algorithm according to the specific data being compressed, optimizing the balance between compression ratio and decompression performance for each data set.

Inventive Principle:

Principle #15Dynamics

2Productivity

If compression method is selected based on past decompression history, then compression efficiency is improved, but prediction accuracy deteriorates for data with varying attributes

Engineering Contradiction:

Improvecompression efficiencyVSAvoiddecompression time prediction accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent performs preliminary analysis of the compressed data by counting code categories before actual decompression. This preliminary action provides accurate information about the decompression time characteristics of the specific data, enabling more precise algorithm selection than relying solely on historical averages.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces the mechanical approach of using historical decomposition data for prediction with a method that directly analyzes the actual compressed data's code category distribution. This substitution provides more accurate predictions by examining the specific characteristics of the data at hand rather than relying on general historical patterns.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If lossless compression method is used to ensure data accuracy, then data integrity is maintained, but compression ratio deteriorates compared to lossy methods

Engineering Contradiction:

Improvedata integrityVSAvoidcompression ratio

Core Design Contradiction:

ReliabilityVSLoss of substance

Solution Approach 1:

The patent applies different compression strategies to different portions of data based on local characteristics. By analyzing code category distributions in the compressed data, the system can identify regions where lossless compression is necessary and regions where more aggressive compression might be acceptable, optimizing the overall balance between integrity and compression ratio.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS11829600B2Storage system and data processing method using data compression

Publication Date: 2023.11.28 HITACHI VANTARA LTD

AI summary

A storage system includes an interface and a data compression system configured to compress reception data from the interface before the data is stored in a storage device. The data compression system is configured to compress the reception data using a first compression algorithm to generate first compressed data, use the number of appearances of each of predetermined code categories included in the first compressed data to estimate a decompression time when a second compression algorithm is used, select a second compression method including compression using the second compression algorithm when the decompression time is equal to or less than a threshold value, and select a first compression method that does not include the compression using the second compression algorithm when the decompression time is greater than the threshold value.