Copy-Redirect on Write for Point-in-Time Storage Volumes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing point-in-time copy technologies in computer storage systems face inefficiencies in managing multiple storage tiers and I/O operations, particularly when creating snapshots and clones, leading to increased metadata usage and storage allocation issues.
Innovation Solution
A method that combines redirect-on-write (ROW) and copy-on-write (COW) algorithms to create snapshots, allowing for efficient write I/O operations within a single storage pool while utilizing multiple storage tiers as needed, by swapping host portions of source volumes with snapshot and clone point-in-time copy volumes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If copy-on-write procedure is used to create snapshot copies, then storage space is reduced through thin provisioning, but the target disk remains dependent on the source disk and I/O operations are impacted
Solution Approach 1:
The patent segments the storage system into multiple independent target disks (first target disk and second target disk) that can be separately managed. Each target disk can maintain its own copy of the source data, allowing independent I/O operations on each disk without affecting the other, thus resolving the dependency issue while maintaining storage efficiency
Solution Approach 2:
The patent introduces a controller as an intermediary that manages the copy operations between the source disk and multiple target disks. The controller coordinates the writing of data to different target disks, enabling independent access and management of each target disk while maintaining the snapshot functionality
2Productivity
If redirect-on-write algorithm is used, then write I/O efficiency is improved within single storage pool, but multiple storage tiers cannot be utilized
Solution Approach 1:
The patent makes the storage system universal by allowing target disks to serve multiple functions: they can be accessed independently for I/O operations, used for creating multiple snapshot copies, and managed with different storage policies. The system can handle both thin-provisioned snapshots and full clones, providing adaptability across different storage scenarios and tiers
3Reliability
If multiple snapshot copies are created, then data protection and analysis capabilities are enhanced, but metadata usage and storage allocation complexity increase
Solution Approach 1:
The patent segments the snapshot management by creating multiple independent target disks instead of managing multiple snapshots on a single disk. Each target disk maintains its own data structure and metadata, simplifying the management complexity while enabling multiple simultaneous snapshot copies with different retention and access characteristics
Data Source
AI summary
A system, method, and computer program product for managing storage volumes in a point-in-time copy cascade. A processor swaps a host portion of a source volume with a host portion of a snapshot point-in-time copy volume. Responsive to an I/O request to overwrite a first data value in a grain of the source volume with a second data value, a processor writes the second data value in a corresponding grain of the snapshot point-in-time copy volume. Responsive to a corresponding grain of a clone point-in-time copy volume not comprising the first data value, a processor copies the first data value to the corresponding grain of the clone point-in-time copy volume.