Storage Data Migration via Logical Mapping Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data migration in storage systems requires manual copying of entire data sets, leading to inefficiencies and prolonged downtime, as each slice of a logical unit must be copied to a new storage pool before it can be accessed, which hinders the ability to notify hosts of migration completion without completing the data transfer.
Innovation Solution
Implementing a method that allows data migration by mapping logical objects from a source storage pool to a destination pool, updating mappings to redirect data access, and enabling hosts to access the destination pool as soon as mappings are updated, thus allowing notification of migration completion without full data transfer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual copying of entire data sets is performed during data migration, then data integrity is ensured, but migration time and system downtime are prolonged
Solution Approach 1:
The patent applies preliminary action by updating the mapping from the source logical unit to the destination logical unit before the actual data copy is complete. This allows the host to be notified of migration completion in advance, reducing downtime while maintaining data integrity through the mapping update mechanism.
2Reliability
If entire data sets are copied before notification of migration completion, then data consistency is maintained, but host downtime is extended
Solution Approach 1:
The mapping update is performed as a preliminary action before the data copy completes. This enables the host to be notified of migration completion and switch to the destination logical unit earlier, improving host availability while data consistency is maintained through the mapping redirection mechanism.
3Productivity
If slice-by-slice copying is performed, then storage space is efficiently utilized, but migration complexity increases
Solution Approach 1:
The patent applies segmentation by copying data slice-by-slice from the source logical unit to the destination logical unit rather than copying the entire data set at once. This improves storage utilization by allowing incremental data transfer while the mapping update mechanism manages the complexity of the segmented migration process.
Data Source
AI summary
A method is used in managing data migration in storage systems. A request to migrate data of a logical object from a source storage pool to a destination storage pool is received. The data of the logical object is mapped to a mapped logical object created in the destination storage pool. The data of the logical object is relocated to storage space allocated in the destination storage pool. Mappings of the mapped logical object are updated to refer to the data relocated to the storage space allocated in the destination storage pool.