Automated Data Storage Tiering Based on Access Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage systems face challenges in determining which data to store on which physical devices to maximize performance, as they lack automated techniques for evaluating and migrating data across different storage tiers based on changing workloads and performance characteristics.
Innovation Solution
A method is introduced to select devices for data storage, determine whether data should be located in a higher performance flash memory tier or a lower performance tier, and automatically move data between tiers based on calculated scores and thresholds, considering read miss density, write percentage, and priority, to optimize storage capacity and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data is stored on higher performance storage tier (flash memory), then I/O throughput is improved, but storage cost increases
Solution Approach 1:
The storage system is segmented into multiple performance tiers (flash memory tier and disk tier), allowing different data to be stored on different tiers based on their access patterns and performance requirements, thereby optimizing the balance between performance and capacity
Solution Approach 2:
Different quality levels of storage are assigned to different data based on their characteristics - high-performance flash memory is allocated to data requiring high I/O throughput, while cost-effective disk storage is used for data with lower access requirements
2Productivity
If automated data evaluation and migration techniques are implemented, then storage performance is optimized, but system complexity increases
Solution Approach 1:
The system automatically evaluates data characteristics and migrates data between tiers without requiring manual intervention, with the automated technique handling the complexity internally while presenting a simplified interface to users
Solution Approach 2:
The system continuously monitors data access patterns and performance metrics, using this feedback to dynamically adjust data placement decisions and optimize storage performance over time
3Productivity
If data migration between storage tiers is performed frequently, then data is optimally positioned, but migration overhead increases
Solution Approach 1:
The system performs data evaluation and migration at periodic intervals rather than continuously, balancing the need to maintain optimal data positioning with the cost of migration operations
Solution Approach 2:
The system proactively migrates data to appropriate tiers based on predicted access patterns and performance requirements, preventing suboptimal performance before it occurs rather than reacting to performance degradation
Data Source
AI summary
Described are techniques for configuring data storage. A plurality of devices is selected. It is determined, for each of the plurality of devices, whether data for said each device should be located in a first storage tier or another storage tier. The first storage tier is higher performance storage tier than the other storage tier. For each of the plurality of devices that the determining step determines should be located in the first storage tier, data of said each device is automatically located on a physical storage device of the first storage tier.