Automated Page-Based Tier Management in Storage Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated page-based tier management in storage systems often fails to achieve optimal data relocation due to ignoring actual usage patterns by user applications, leading to suboptimal storage resource utilization.
Innovation Solution
A computerized storage system with a data migration module that automatically migrates data between tier segments based on usage patterns, allowing for dynamic adjustment of migration conditions to align with the host computer's data usage policies, enabling the storage system to optimize data placement according to actual application needs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If automated page-based tier management is performed by the storage system itself, then data relocation can be performed automatically based on access characteristics, but the actual usage of data by user applications is ignored leading to suboptimal relocation results
Solution Approach 1:
The patent introduces an intermediary component (file server or host computer) that bridges the storage system's automated tier management and the application's data usage knowledge. This intermediary receives data usage information from applications and translates it into migration condition settings that the storage system can execute, allowing automated operation while incorporating application-awareness.
Solution Approach 2:
The patent makes the migration conditions dynamic by allowing them to be adjusted based on actual application data usage patterns. Instead of fixed automated rules, the system dynamically modifies migration thresholds, triggers, and parameters according to real-world usage feedback, enabling both automation and adaptability.
2Ease of operation
If data migration is performed automatically without considering application usage patterns, then operation simplicity is improved, but storage resource utilization efficiency deteriorates
Solution Approach 1:
The system enables self-service by allowing applications to provide their own usage metadata and access patterns to the storage system. Applications essentially serve themselves by contributing knowledge about their data, which the storage system then uses to optimize migration decisions, maintaining simplicity while improving efficiency.
Solution Approach 2:
The patent implements feedback loops where application data usage patterns are continuously monitored and fed back to adjust migration conditions. This feedback mechanism allows the system to learn from actual usage and improve storage resource utilization over time while maintaining automated operation.
Data Source
AI summary
The system includes host computers, file servers and a storage system having automated page-based management means. The storage system interface receives instructions to change the condition for decision for migration regarding particular parts or the whole volume. The host computer can control execution of the migration performed by the storage system by specifying areas or volumes with the condition via the interface. Highly optimized, appropriate data placement and data relocation in computer system can be achieved when the application, host computer or management computer can recognize or predict the usage of the data or files. The storage system having automated page-based management may include compression/decompression and a control method for the compression and decompression process.