Block Storage Compression Using Adaptive Parameter Tables

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Memory systems with small access granularities, such as SSDs and CXL memories, face challenges in achieving high compression ratios due to the small size of data units processed, making it difficult to effectively compress data.

Innovation Solution

Implementing a compression/decompression circuit that utilizes a compression parameter table to adaptively calculate and apply compression parameters based on recent data patterns, allowing for high compression ratios without reducing throughput.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If data compression is performed on small access granularity units (4 KB or less), then the data can be processed by memory systems with small access granularities, but the compression ratio becomes low

Engineering Contradiction:

Improveadaptability to small access granularityVSAvoidcompression ratio

Core Design Contradiction:

Adaptability or versatilityVSLoss of substance

Solution Approach 1:

The patent merges multiple small data units (4 KB or less) into a larger combined data unit for compression processing. By combining several small access granularity units together, the compression device can process a larger data volume that achieves better compression ratios, while still being compatible with memory systems that have small access granularities.

Inventive Principle:

Principle #5Merging (Combining)

2Loss of substance

If compression parameters are determined for each data to be compressed, then high compression ratio can be achieved, but the device complexity increases

Engineering Contradiction:

Improvecompression ratioVSAvoidcompression parameter management

Core Design Contradiction:

Loss of substanceVSDevice complexity

Solution Approach 1:

The patent pre-calculates and stores compression parameters for multiple types of data before actual compression occurs. A compression parameter table is prepared in advance, containing pre-determined compression parameters for different data types. When compression is needed, the appropriate pre-calculated parameters are simply retrieved from the table, avoiding the need for complex real-time parameter calculation and reducing device complexity.

Inventive Principle:

Principle #10Preliminary action

3Loss of substance

If initial dictionary is tailored to specific application, then compression ratio improves, but the ease of operation decreases for general-purpose data

Engineering Contradiction:

Improvecompression ratioVSAvoidgeneral-purpose data processing

Core Design Contradiction:

Loss of substanceVSEase of operation

Solution Approach 1:

The patent creates a universal compression system that can handle multiple data types effectively. Instead of requiring separate tailored dictionaries for each application, the system uses a compression parameter table that contains pre-calculated parameters suitable for various data types. This universal approach allows the same compression device to achieve good compression ratios across different applications and data types without requiring application-specific configuration.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12572276B2Data compression methods for block-based storage systems

Publication Date: 2026.03.10 KIOXIA CORP

AI summary

According to one embodiment, a memory controller includes a parameter table having entries which respectively correspond to superblock address ranges. The memory controller translates a logical address of data to be written to the memory into a superblock address, calculates compression parameters which correspond to items of data to be written to superblock addresses, writes the compression parameters to the parameter table, and compress data using the parameter table.