Storage Subsystem Dynamic Compression Based on Pool Usage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The use of compression techniques in storage systems can lead to deteriorated response performance for high-level devices due to CPU load, and existing asynchronous compression methods still result in delayed access times when decompressing data.
Innovation Solution
Implementing a storage subsystem that dynamically adjusts compression based on real pool usage rates, prioritizing compression for virtual volumes when the usage rate exceeds a set threshold and stopping compression when it falls below another threshold, thereby optimizing CPU usage and response times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If compression processing is performed on virtual volumes, then capacity efficiency is improved, but response performance deteriorates due to CPU load
Solution Approach 1:
The patent implements periodic compression processing at predetermined intervals rather than continuously or on-demand. The compression execution determination unit decides whether to perform compression based on timing information, executing compression only at specific periodic moments when it determines compression is appropriate, thus reducing CPU load while maintaining capacity efficiency benefits
Solution Approach 2:
The patent dynamically adjusts compression execution based on real-time pool usage rates and access patterns. The compression execution determination unit monitors the usage rate of the real pool and access frequencies of virtual volumes, adapting the compression strategy dynamically to current system conditions, thereby optimizing the balance between capacity efficiency and response performance
2Speed
If asynchronous compression is used to reduce CPU load, then response performance is improved, but access time is delayed due to decompression requirements
Solution Approach 1:
The patent performs compression processing in advance during periodic intervals before data access occurs. By proactively compressing data during low-utilization periodic windows, the system prepares compressed data ahead of time, eliminating the need for time-consuming decompression during actual access operations
Solution Approach 2:
The patent creates and manages multiple copies of data in different compression states. The system maintains both compressed and uncompressed versions of data, allowing it to serve read requests directly from uncompressed copies when needed, thus avoiding decompression delays while preserving the space benefits of compression
3Quantity of substance
If compression is continuously performed to maximize capacity efficiency, then storage utilization is improved, but CPU resource consumption increases
Solution Approach 1:
The patent replaces continuous compression with periodic compression executed at predetermined intervals. The compression execution determination unit evaluates whether compression should occur at each periodic interval based on current pool usage rates and access patterns, significantly reducing CPU resource consumption compared to continuous compression while maintaining effective storage utilization
Solution Approach 2:
The patent changes the operational parameters of compression by introducing timing-based control and usage-rate-based thresholds. Instead of constant compression, the system adjusts compression execution based on temporal parameters (intervals) and usage parameters (pool usage rates), optimizing the balance between storage utilization and CPU consumption
Data Source
AI summary
The present invention relates to executing data compression based on the used capacities of real pools. The present storage subsystem stores a definition information regarding conditions of starting data compression and ending data compression defined via usage rates of a real pool as physical area, wherein if the usage rate of the real pool satisfies the condition for starting compression, the compression processing is started, and if the usage rate satisfies the condition for ending compression, the compression processing is ended.