Host Data Compression Logic for Storage Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data storage systems face a conflict between achieving high data compression and maintaining high throughput in processing host I/O requests, as high compression levels are costly in terms of CPU utilization and can hinder rapid processing.
Innovation Solution
The technology identifies infrequently accessed host data and compresses it using secondary compression logic to generate a highly compressed version, storing it in a lower performance tier of non-volatile data storage, while maintaining initial compression for frequently accessed data to ensure rapid processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of substance
If high level data compression is performed on host data during background processing, then data reduction level is improved, but device complexity increases due to multiple compression logic components
Solution Approach 1:
The patent divides the compression system into multiple independent compression logic components (first compression logic for initial compression, second compression logic for deep compression). Each compression logic operates independently on different data sets based on access patterns, allowing the system to achieve high overall compression while maintaining modularity and manageability.
Solution Approach 2:
The patent applies different compression strategies to different portions of data based on their access characteristics. Frequently accessed data in hot data sets receives initial compression only, while infrequently accessed data in cold data sets undergoes both initial and deep compression. This local differentiation optimizes the balance between compression ratio and system complexity.
2Loss of substance
If deep compression is applied to all host data, then data reduction level is improved, but processing throughput deteriorates due to increased CPU utilization
Solution Approach 1:
The system performs preliminary classification of data into hot and cold data sets based on access patterns before applying deep compression. By identifying infrequently accessed data in advance, the system can apply intensive deep compression only to appropriate data sets, avoiding the throughput penalty that would result from compressing all data equally.
Solution Approach 2:
Instead of applying deep compression uniformly to all data, the patent applies it partially only to cold data sets where the compression benefits outweigh the processing costs. This selective application achieves significant overall data reduction while maintaining high throughput for frequently accessed data.
Data Source
AI summary
Host data stored in one or more source physical extents of non-volatile data storage is identified as valid and determined to be infrequently written by host I/O requests, and is therefore compressed to generate a highly compressed version of the valid host data. The highly compressed version is then stored into at least one target physical extent. The valid host data may be initially compressed before it is stored in the source physical extent(s), and may be re-compressed to generate the highly compressed version. If the valid host data is also infrequently read, it may be recompressed using larger blocks of host data than were used to perform the initial compression. The performance tier of the target physical extent may be different from (e.g. lower than) the performance tier of the source physical extent. The technology may be embodied in a background process such as a garbage collector.