Storage Area Compression Using IO-Based Access Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The frequent compression and decompression of merged data blocks in storage systems lead to increased overhead and performance degradation due to decompression, data block movement, reorganization, and metadata updates, especially when access patterns change frequently.
Innovation Solution
A method that determines access levels of storage areas based on historical IO counts, sets a dynamic access level threshold, and selectively compresses pages in target storage areas to minimize unnecessary operations and optimize storage space usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple data blocks are merged and compressed together to improve compression ratio, then storage space efficiency is improved, but system overhead increases due to decompression, data block movement, reorganization, and metadata updates
Solution Approach 1:
The patent implements dynamic access level thresholds that adjust based on current storage system state. The threshold is not fixed but changes according to IO workload patterns and storage utilization, allowing the system to adaptively determine which storage areas qualify for compression. This resolves the contradiction by making compression decisions dynamic rather than static, optimizing the balance between storage efficiency and operational overhead.
Solution Approach 2:
The patent changes the parameter of access level threshold from a static value to a dynamically adjustable parameter based on IO counts and storage utilization. By monitoring historical IO patterns and adjusting the threshold parameter accordingly, the system can optimize compression decisions to achieve better storage efficiency while minimizing unnecessary decompression and reorganization operations.
2Quantity of substance
If data blocks are frequently compressed and decompressed to maintain optimal storage utilization, then storage capacity management is improved, but performance degradation occurs due to repeated decompression and recompression operations
Solution Approach 1:
The patent performs preliminary analysis of IO counts and access patterns before making compression decisions. By evaluating historical IO data and predicting future access patterns, the system pre-determines which storage areas are suitable for compression, avoiding frequent decompression/recompression of actively accessed data. This preliminary action prevents performance degradation by ensuring compression only occurs when appropriate.
Solution Approach 2:
The patent implements a feedback mechanism that continuously monitors IO counts and storage utilization, using this information to adjust compression decisions. The system learns from past compression outcomes and IO patterns, refining its access level threshold to minimize unnecessary decompression operations. This feedback loop maintains storage capacity optimization while preserving performance by avoiding redundant compression cycles.
3Ease of operation
If a fixed compression policy is applied to all storage areas to simplify management, then operational simplicity is improved, but adaptability to different access patterns deteriorates
Solution Approach 1:
The patent segments the storage space into multiple storage areas and applies different compression decisions to each segment based on its specific IO characteristics. Rather than a uniform policy, each storage area is evaluated independently using its own IO count history, allowing the system to adapt to diverse access patterns across different regions of storage while maintaining manageable complexity through automated segment-level analysis.
Data Source
AI summary
Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for compressing data. The method includes determining a plurality of access levels of a plurality of storage areas in a storage space based on a plurality of input/output (IO) counts of the plurality of storage areas in a historical time period. The method further includes determining an access level threshold based on the utilization of the storage space. The method further includes selecting at least one target storage area among the plurality of storage areas based on the plurality of access levels and the access level threshold. The method further includes compressing a plurality of pages in the at least one target storage area. Embodiments of the present disclosure provide a strategy of merging and compression based on a system state, which balances data compression efficiency and system performance.