Fixed-Length Compressed Blocks for Parallel Decompression

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The inefficiency of data decompression in hardware chips due to varying compression rates of raw data blocks and the need for synchronized decompression by multiple engines, leading to idle computation capabilities and reduced algorithm running speed.

Innovation Solution

Standardizing the compression process to generate fixed-length standard compressed data blocks, allowing multiple decompression engines to operate in parallel, ensuring each block can be decompressed within one processing cycle.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If multiple decompression engines are used for parallel decompression, then decompression throughput is improved, but decompression efficiency is reduced due to varying processing times for different compressed data blocks

Engineering Contradiction:
Improvedecompression throughputVSAvoiddecompression efficiency
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The original compressed data is segmented into multiple equal-length standard compressed data blocks, each of which can be independently decompressed by a separate decompression engine. This segmentation enables parallel processing while maintaining uniform processing time for each block, thus improving throughput without sacrificing efficiency.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The invention changes the parameter of compressed data block length from variable to fixed/equal length. By standardizing the length of compressed data blocks, all blocks can be processed in the same time by decompression engines, eliminating the idle time caused by waiting for the slowest block to complete.

Inventive Principle:
Principle #35Parameter changes

2Loss of substance

If variable-length compressed data blocks are used, then compression ratio is optimized, but parallel decompression efficiency is reduced due to synchronization requirements

Engineering Contradiction:
Improvedata volumeVSAvoidparallel decompression efficiency
Core Design Contradiction:
Loss of substanceVSProductivity

Solution Approach 1:

The compressed data is divided into multiple standard blocks of equal length, where each block is a self-contained unit that can be independently processed. This segmentation allows multiple decompression engines to work in parallel without requiring complex synchronization mechanisms, as all blocks complete processing in the same time.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Instead of adapting the decompression process to handle variable-length blocks, the invention inverts the approach by standardizing the block length to match the fixed processing capacity of decompression engines. This inversion simplifies the parallel processing architecture and eliminates synchronization overhead.

Inventive Principle:
Principle #13The other way round (Inversion)

Data Source

PatentEP3820048B1Data compression and decompression method and related apparatus, electronic device, and system
Publication Date: 2025.07.30 HUAWEI TECH CO LTD
  • EP3820048B1 patent drawingFigure 1
  • EP3820048B1 patent drawingFigure 2
  • EP3820048B1 patent drawingFigure 3

AI summary

Embodiments of this application disclose a data compression method, a data decompression method, and a related apparatus, an electronic device, and a system. In the data compression method, a raw data block in raw data is processed based on a compression algorithm to obtain a standard compressed data block that has a length of L2 and that corresponds to the raw data block, and the raw data is further compressed into one or more standard compressed data blocks that each have the length of L2 and that are to be decompressed in parallel by a decompression apparatus, where the decompression apparatus includes a plurality of decompression engines, and each decompression engine is capable of decompressing one standard compressed data block within one processing cycle. According to the data compression method in the embodiments of the present invention, a standard compressed data block with a fixed length can be obtained through compression. In this way, various decompression engines can synchronously decompress a plurality of standard compressed data blocks. This improves data decompression efficiency.