Data Segment Grouping for Similar-Property Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional storage systems face inefficiencies in data compression due to mixing dissimilar data types within the same data segments, which reduces the identification of redundancies and thus the efficiency of compression operations.
Innovation Solution
The storage system identifies and groups data blocks with similar properties together within data segments, allowing for more efficient redundancy identification and compression by reading only a portion of the data to determine properties, such as the top bits, to categorize and store similar data types in proximity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If dissimilar data types are mixed within the same data segment, then storage capacity is utilized, but compression efficiency deteriorates due to reduced redundancy identification
Solution Approach 1:
The patent divides data segments into sub-segments based on data properties (e.g., compressible vs. non-compressible). This segmentation allows similar data types to be grouped together within each sub-segment, enabling effective redundancy identification and compression while maintaining overall storage capacity utilization.
Solution Approach 2:
The patent applies different quality characteristics to different parts of the data segment by identifying specific data properties (such as compressibility) and treating similar data types with the same compression algorithms. This local quality approach ensures that each data type receives appropriate processing, maximizing compression efficiency without sacrificing storage capacity.
2Productivity
If data blocks are grouped by similar properties, then compression efficiency is improved, but device complexity increases due to property identification and categorization
Solution Approach 1:
The patent performs property identification on only a portion of each data block (e.g., reading only the first few bytes or specific metadata fields) rather than analyzing the entire data block. This partial action approach significantly reduces the computational overhead and device complexity while still enabling effective data categorization and grouping for compression.
Solution Approach 2:
The patent uses lightweight, inexpensive property identification methods that require minimal processing resources. By using simple criteria (such as checking data type headers or first few bytes) rather than complex analysis, the system achieves effective data grouping with minimal additional device complexity.
3Productivity
If only a portion of data is read to determine properties, then computing overhead is minimized, but measurement precision may be insufficient for accurate categorization
Solution Approach 1:
The patent performs preliminary property identification by reading only the essential portion of data (such as headers, metadata, or first few bytes) that contain sufficient information for accurate categorization. This preliminary action provides enough precision for effective grouping while minimizing computing overhead, as the critical property information is typically located in the initial portion of data blocks.
Data Source
AI summary
A plurality of data blocks stored at a plurality of solid-state storage devices are identified. A portion of data is read from each data block of the plurality of data blocks. A corresponding property is determined for each data block of the plurality of data blocks based on reading the portion of the data. A set of data blocks from the plurality of data blocks is identified, wherein each data block of the set of data blocks is associated with a first corresponding property. The set of data blocks is stored at a data segment.