Segmented Data Compression Layout for SSD Page Throughput
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current memory systems face limitations in increasing throughput and storage capacity due to inefficiencies in data compression and management, particularly in solid state drives (SSDs), where existing methods do not effectively optimize the compression of data segments across pages, leading to suboptimal use of storage space and transfer efficiency.
Innovation Solution
The proposed solution involves a method for data compression and management that determines the compressibility of each data segment, forms compressed managed units (CMUs) with varying compression levels, and organizes these units within pages to maximize storage efficiency, using a controller with compression/decompression components and metadata to manage the physical organization and compression values of CMUs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data compression is applied to increase storage capacity, then storage efficiency improves, but data access time and processing complexity increase
Solution Approach 1:
The patent divides data into managed units (MUs) that are processed and compressed independently. Each MU is a fixed-size block that can be compressed separately, allowing parallel processing and reducing overall compression time. The segmentation enables the system to handle large datasets efficiently by breaking them into manageable chunks that can be compressed and stored independently.
Solution Approach 2:
The patent performs compression on entire managed units before storage, rather than compressing individual data segments during access. This preliminary compression action stores data in a ready-to-use compressed format, eliminating the need for real-time compression during read operations and reducing data access time.
2Quantity of substance
If uniform compression is applied to all data segments, then processing simplicity is maintained, but storage efficiency decreases due to inability to optimize for varying compressibility
Solution Approach 1:
The patent applies different compression levels to different managed units based on their individual compressibility characteristics. Each MU is evaluated separately and assigned an appropriate compression level (0-3), allowing the system to optimize storage efficiency for each unit while maintaining manageable processing complexity through standardized compression levels.
Solution Approach 2:
The patent changes the compression parameter (compression level) dynamically based on the measured compressibility of each managed unit. By adjusting the compression level from 0 to 3 according to actual data characteristics, the system achieves optimal storage efficiency without applying excessive compression to already-compressible data or insufficient compression to highly compressible data.
3Quantity of substance
If compression levels are increased to maximize storage density, then storage capacity improves, but data transfer speed and access performance deteriorate
Solution Approach 1:
The patent applies compression selectively at partial levels rather than maximum compression for all data. By offering compression levels 0-3, the system can choose the appropriate degree of compression based on data characteristics and performance requirements, avoiding the performance penalty of excessive compression while still achieving good storage density for suitable data types.
Data Source
AI summary
The present disclosure includes apparatuses and methods for data compression and management. A number of methods include receiving a number of data segments corresponding to a managed unit amount of data, determining a respective compressibility of each of the number of data segments, compressing each of the number of data segments in accordance with its respective determined compressibility, forming a compressed managed unit that includes compressed and/or uncompressed data segments corresponding to the number of data segments corresponding to the managed unit amount of data, and forming a page of data that comprises at least the compressed managed unit.