Parallel Data Migration Control for Storage Hierarchical Layers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data migration processes in storage hierarchical control systems are inefficient due to the lack of effective control mechanisms to reduce the overall time period for executing data migration between storage layers, leading to suboptimal access performance.
Innovation Solution
An apparatus that identifies candidate unit regions for data migration, determines if data migration is being executed on divided regions, and instructs parallel migration to optimize the process by avoiding waiting periods and utilizing available resources efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If data migration is executed sequentially without parallel processing, then the migration process is simple to manage, but the overall data migration time is extended
Solution Approach 1:
The patent divides the physical storage region into multiple divided regions and divides data migration into multiple units, allowing parallel execution of migration operations on different divided regions simultaneously, thereby reducing overall migration time while maintaining manageable complexity through structured organization
Solution Approach 2:
The patent introduces a new dimension of parallel processing by executing data migration operations across multiple divided regions concurrently rather than sequentially, transforming the migration process from a single-threaded sequential operation to a multi-threaded parallel operation, thereby significantly reducing migration time
2Loss of time
If data migration is executed in parallel on multiple divided regions, then the overall data migration time is reduced, but the control complexity increases
Solution Approach 1:
The patent implements a control mechanism that monitors the execution status of data migration on each divided region and dynamically adjusts migration instructions based on real-time feedback, allowing parallel execution while maintaining controlled complexity through status-aware decision making
Solution Approach 2:
The patent employs dynamic control where the system adaptively determines whether to execute parallel migration or sequential migration based on current system state and resource availability, transforming the rigid control structure into a flexible dynamic control mechanism that optimizes performance without excessive complexity
3Reliability
If data migration waits for completion on divided regions, then data consistency is maintained, but idle periods increase and efficiency decreases
Solution Approach 1:
The patent ensures continuous useful action by overlapping data migration operations across multiple divided regions, allowing the system to continuously perform migration work without idle waiting periods, thereby maintaining both data consistency through proper coordination and high efficiency through uninterrupted parallel execution
Data Source
AI summary
In a state where one of physical storage regions is allocated to each of unit regions included in a logical storage region, an apparatus identifies, from among the unit regions, first unit regions which have been allocated from physical storage regions whose allocations to the first unit regions are to be changed. The apparatus sequentially selects, from among the first unit regions, a second unit region as a candidate for data migration, and instructs data migration of the second unit region between the physical storage regions. The apparatus determines whether data migration of any one of the unit regions whose data has been instructed to be migrated is being executed on each of multiple divided regions obtained by dividing a physical storage region to which data of the second unit region is to be migrated, and instructs data migration of the second unit region based on the determination.