Data Migration via Slice Activity Subsets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data storage systems face performance constraints due to uneven distribution of load across disks and limited working memory capacity, which affects data migration efficiency across different tiers in storage arrays.
Innovation Solution
A method is implemented to identify slice activity data, divide it into subsets, store and update subsets in working memory, and periodically copy them to disk drives, allowing for efficient data migration by categorizing data as hot or cold and relocating it across tiers based on access frequency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all slice activity data is stored in working memory, then data migration decisions can be made with complete information, but working memory capacity is exceeded
Solution Approach 1:
The patent divides slice activity data into multiple subsets, each stored in a separate data structure within working memory. This segmentation allows the system to manage activity information for a portion of slices at a time, fitting within limited working memory capacity while still maintaining comprehensive tracking across all slices through systematic rotation and updating of these subsets.
2Productivity
If data is frequently accessed, then user performance is improved, but load distribution across disks becomes uneven
Solution Approach 1:
The patent implements periodic data migration by monitoring slice activity over time intervals and relocating data between tiers based on accumulated activity patterns. This periodic reassessment and migration approach balances load distribution across disks while ensuring frequently accessed data remains on high-performance tiers, thereby maintaining user performance without sustained uneven loading.
3Speed
If data is migrated based on real-time activity, then response time is optimized, but system complexity increases
Solution Approach 1:
The patent pre-allocates data structures for storing slice activity information and establishes predetermined thresholds for data migration. By preparing these structures and criteria in advance, the system can make rapid migration decisions based on pre-defined rules rather than complex real-time analysis, thus optimizing response time while controlling system complexity through structured, rule-based operation.
Data Source
AI summary
A method is used in managing data activity information for data migration in data storage systems. Slice activity data for slices stored in a data storage system is identified. The slice activity data is divided into multiple subsets of slice activity data. A subset of slice activity data may be obtained and stored in working memory. The slice activity data may be updated to correspond to changes slice activity. After a determined time, the updated slice activity data is copied to a disk drive and the method is repeated with the next subset of slice activity data.